Compare commits

..

24 Commits

Author SHA1 Message Date
lebaudantoine de50aeb4fe wip hint if the user is authenticated or not in attributes 2026-05-17 23:44:49 +02:00
lebaudantoine 71f76a81e9 ♻️(backend) refactor caller identity getter
Enhance getting the caller's identity to prevent None.
2026-05-17 23:42:54 +02:00
lebaudantoine 385da86759 🔒️(backend) verify participant presence before mute operations
Ensure the participant requesting a mute action is still present
in the room before processing the request.

This mitigates scenarios where a previously issued token could be
reused after the meeting has ended.

Current token lifetime is intentionally long-lived and will be
refactored in the future to better align with LiveKit session
constraints. In the meantime, add this extra validation step to
reduce the attack surface.
2026-05-17 23:39:53 +02:00
lebaudantoine 81e3483f28 📝(changelog) update the changelog 2026-05-17 23:39:53 +02:00
lebaudantoine 5e030c2a07 ♻️(frontend) refactor useMuteParticipant hook
Fix unreachable code when notifying participants that they were
muted.

Prevent unnecessary function re-creations when props remain
unchanged.

Also guard against missing tokens by logging an error and
returning early when the token is undefined.
2026-05-17 23:39:53 +02:00
lebaudantoine 32fbedd358 (backend) extend live synchronization to lobby access level updates
Extend the existing live synchronization mechanism beyond room
configuration to also include lobby access level changes.

This ensures that all owners and admins sharing a room maintain a
consistent and up-to-date view of room state in the frontend,
including configuration and access control updates.
2026-05-17 23:39:53 +02:00
lebaudantoine aab90650f1 (frontend) add synchroniser for room metadata updates
Listen to room metadata change events and synchronize the React
Query cache with the latest room data fetched from the API.

This ensures clients react to live configuration updates, such as
showing or hiding mute controls when `everyone_can_mute` changes.
2026-05-17 23:39:53 +02:00
lebaudantoine 534cf000b2 (backend) expose room configuration to all API consumers
Update room serialization to include room configuration for all
users fetching the API response, not only room owners.
This behavior was inherited from the original upstream project.

At the moment, exposing this configuration does not appear to
introduce meaningful security concerns or provide attackers with
additional capabilities.

The decision will continue to be reviewed from a security
perspective, but sharing the configuration improves frontend
consistency and synchronization.
2026-05-17 23:39:53 +02:00
lebaudantoine 5bac1668fe ♻️(fullstack) simplify source serialization
Simplify source serialization and validation logic while improving
type safety around room configuration handling.

Introduce a dedicated TypeScript type matching the backend
Pydantic model more precisely.

Also harmonize track source casing between frontend and backend to
remove redundant conversion logic and resolve #1282.
2026-05-17 23:39:53 +02:00
lebaudantoine 5a7a0da923 (backend) add synchronization mechanism for room configuration updates
Introduce synchronization of room configuration changes across
active participants.

When a room configuration is updated through a PUT operation, the
backend now performs an additional LiveKit API call to notify room
participants through a room metadata update event.

This ensures admins and owners quickly see up-to-date settings in
their administration panel. It also prepares the frontend for
automatic updates of unprivileged participants room’s data without
refetching it from the API.

An event-driven design was chosen instead of storing the full room
configuration in LiveKit metadata. While embedding the state
directly in metadata would provide immediate synchronization, it
would also require initializing and maintaining configuration
state during room creation or webhook handling, increasing the
risk of operational failures and regressions.

Instead, the backend emits lightweight synchronization events and
active clients update their React Query cache, which remains the
single source of truth for room configuration data.
2026-05-17 23:39:53 +02:00
lebaudantoine c20daafd81 (fullstack) support everyone_can_mute room configuration
Introduce a new room setting controlling whether all participants,
including non-privileged users, can mute others.

Update API validation accordingly and add the frontend controls
allowing administrators to toggle the option and persist the
configuration through the API.
2026-05-17 23:39:53 +02:00
lebaudantoine 9846a61bd0 (frontend) update useCanMute hook to reflect room muting behavior
Allow non-privileged users to mute others when the
everyone_can_mute configuration is unset or true.

This setting is not yet customizable by room owners and will be
introduced in a future update.
2026-05-17 23:39:53 +02:00
lebaudantoine 388b7d172d (frontend) allow unauthenticated participants to mute via LiveKit token
Pass the LiveKit token when calling the mute-participant endpoint
to authenticate the request.

This enables non-authenticated participants to mute others through
the API while preserving proper authorization checks.
2026-05-17 23:39:53 +02:00
lebaudantoine 288562cc0e 🛂(backend) allow participants to mute others based on room configuration
Enable any participant to mute others when the room configuration
allows it. This is enabled by default for all meetings unless
explicitly disabled by an administrator.

Privileged users retain the ability to mute any participant
regardless of the room configuration.
2026-05-17 23:39:53 +02:00
leo 79400188d8 🔊(summary) improve logging of speaker assign
Structure logging of speaker assignment in json format to help
assess its performance.
2026-05-14 15:39:02 +02:00
lebaudantoine dcaa45ccfe 🩹(frontend) fix subtitle background regression
Restore transparent background as the default subtitle background
to match previous behavior.
2026-05-14 15:04:48 +02:00
lebaudantoine 35951ba2a6 🔖(minor) bump release to 1.16.0 2026-05-13 22:30:32 +02:00
lebaudantoine 72184e1370 🩹(frontend) fix spacing regression in mobile control bar
Correct excessive spacing between action buttons in the mobile
control bar introduced by a recent layout change.
2026-05-13 20:15:32 +02:00
leo 1b4a8fbac2 🔧(agents) fix Docker setup
Fix two issues. 1: Missmatch between commands in dev and production in
Dockerfile, leading to unexpected behaviors. 2: Naming of
multi-user-transcriber -> multi-user-transcriber-dev for coherence.
2026-05-13 20:07:45 +02:00
lebaudantoine 1e2fad5444 ️(mail) revert mail upgrade due to unhandled breaking changes
Rollback the mail package upgrade after identifying multiple
breaking changes introduced in v5 that were not fully accounted
for.

Local testing initially missed the issue because the mail Docker
image had not been rebuilt automatically, causing broken emails to
go unnoticed.
2026-05-13 19:55:54 +02:00
leo 96f97ed2d0 (summary) improve speaker assignment
Speaker-to-participant assignment relie on WhisperX word timings, but
incorrect word durations in the output can lead to inaccurate overlap
scoring and wrong user attribution. Add a custom heuristic to trim
overly long word durations before computing assignments.
2026-05-12 16:58:07 +02:00
lebaudantoine 02d16cb55c ⬆️(addons) update dependencies 2026-05-12 16:26:16 +02:00
lebaudantoine 7268ff6777 ⬆️(mail) update dependencies 2026-05-12 16:26:16 +02:00
lebaudantoine cca5bc2186 ⬆️(frontend) update dependencies 2026-05-12 16:26:16 +02:00
50 changed files with 1999 additions and 1381 deletions
+14 -3
View File
@@ -10,6 +10,18 @@ and this project adheres to
### Added
- ✨(fullstack) allow participants to mute others based on room configuration
- ✨(frontend) add synchronizer for room metadata updates
### Changed
- ♻️(fullstack) simplify source serialization
- ✨(backend) expose room configuration to all API consumers
## [1.16.0] - 2026-05-13
### Added
- 🔒️(backend) add validation of Room.configuration
- ✨(helm) add support multiple transcribe worker / endpoint #1247
- ✨(backend) make LiveKit Egress recording encoding configurable #1288
@@ -19,10 +31,8 @@ and this project adheres to
- ♻️(summary) change tasks endpoint signature
- ⬆️(dependencies) update urllib3 to v2.7.0 [SECURITY]
### Changed
- 🧑‍💻(agents) use `uv` for package management
- ✨(summary) improve speaker-to-participant assignment
### Fixed
@@ -32,6 +42,7 @@ and this project adheres to
- ⬆️(mail) fix dependencies not having resolved or integrity field #1321
- 🐛(summary) complete webm support #1328
- 🐛(backend) add link to "Open" text in recording email
- 🩹(frontend) fix spacing regression in mobile control bar
## [1.15.0] - 2026-04-30
+2 -2
View File
@@ -109,7 +109,7 @@ build-frontend: ## build the frontend container
.PHONY: build-frontend
build-agents: ## build the multi-user-transcriber agent container
@$(COMPOSE) build multi-user-transcriber
@$(COMPOSE) build multi-user-transcriber-dev
.PHONY: build-agents
down: ## stop and remove containers, networks, images, and volumes
@@ -138,7 +138,7 @@ run-agents: ## start the multi-user-transcriber agent
.PHONY: run-agents
run-agent-multi-user-transcriber: ## start the LiveKit agents (multi users transcriber)
@$(COMPOSE) up --force-recreate -d multi-user-transcriber
@$(COMPOSE) up --force-recreate -d multi-user-transcriber-dev
.PHONY: run-agent-multi-user-transcriber
run-agent-metadata-collector: ## start the LiveKit agents (metadata collector)
+10 -11
View File
@@ -271,6 +271,16 @@ services:
- action: rebuild
path: ./src/agents
multi-user-transcriber-dev:
build:
context: ./src/agents
target: development
env_file:
- env.d/development/multi_user_transcriber
volumes:
- ./src/agents:/app
- /app/.venv
redis-summary:
image: redis
ports:
@@ -332,17 +342,6 @@ services:
- action: rebuild
path: ./src/summary
multi-user-transcriber:
build:
context: ./src/agents
target: development
command: ["python", "multi_user_transcriber.py", "dev"]
env_file:
- env.d/development/multi_user_transcriber
volumes:
- ./src/agents:/app
- /app/.venv
networks:
default:
resource-server:
+3 -15
View File
@@ -1,21 +1,9 @@
AWS_S3_ENDPOINT_URL=minio:9000
AWS_S3_ACCESS_KEY_ID=meet
AWS_S3_SECRET_ACCESS_KEY=password
LIVEKIT_URL=ws://livekit:7880
LIVEKIT_API_KEY=devkey
LIVEKIT_API_SECRET=secret
STT_PROVIDER=voxtral-vllm # voxtral-vllm, kyutai, deepgram
STT_PROVIDER=kyutai
ENABLE_SILERO_VAD=False
DEEPGRAM_API_KEY=your-deepgram-api-key
KYUTAI_STT_BASE_URL=url
KYUTAI_API_KEY=your-kyutai-api-key
VOXTRAL_VLLM_BASE_URL=wss://<host>/v1/realtime
VOXTRAL_VLLM_MODEL=voxtral-mini-4b-realtime-2602
VOXTRAL_VLLM_API_KEY=your-vllm-api-key
VOXTRAL_VLLM_TARGET_STREAMING_DELAY_MS=480
KYUTAI_STT_BASE_URL=
KYUTAI_API_KEY=
+12 -250
View File
@@ -19,7 +19,7 @@
"@types/office-runtime": "^1.0.35",
"acorn": "^8.11.3",
"babel-loader": "^9.1.3",
"copy-webpack-plugin": "^12.0.2",
"copy-webpack-plugin": "^14.0.0",
"eslint-plugin-office-addins": "^4.0.3",
"file-loader": "^6.2.0",
"html-loader": "^5.0.0",
@@ -4319,44 +4319,6 @@
"node": ">= 4.0.0"
}
},
"node_modules/@nodelib/fs.scandir": {
"version": "2.1.5",
"resolved": "https://registry.npmjs.org/@nodelib/fs.scandir/-/fs.scandir-2.1.5.tgz",
"integrity": "sha512-vq24Bq3ym5HEQm2NKCr3yXDwjc7vTsEThRDnkp2DK9p1uqLR+DHurm/NOTo0KG7HYHU7eppKZj3MyqYuMBf62g==",
"dev": true,
"license": "MIT",
"dependencies": {
"@nodelib/fs.stat": "2.0.5",
"run-parallel": "^1.1.9"
},
"engines": {
"node": ">= 8"
}
},
"node_modules/@nodelib/fs.stat": {
"version": "2.0.5",
"resolved": "https://registry.npmjs.org/@nodelib/fs.stat/-/fs.stat-2.0.5.tgz",
"integrity": "sha512-RkhPPp2zrqDAQA/2jNhnztcPAlv64XdhIp7a7454A5ovI7Bukxgt7MX7udwAu3zg1DcpPU0rz3VV1SeaqvY4+A==",
"dev": true,
"license": "MIT",
"engines": {
"node": ">= 8"
}
},
"node_modules/@nodelib/fs.walk": {
"version": "1.2.8",
"resolved": "https://registry.npmjs.org/@nodelib/fs.walk/-/fs.walk-1.2.8.tgz",
"integrity": "sha512-oGB+UxlgWcgQkgwo8GcEGwemoTFt3FIO9ababBmaGwXIoBKZ+GTy0pP185beGg7Llih/NSHSV2XAs1lnznocSg==",
"dev": true,
"license": "MIT",
"dependencies": {
"@nodelib/fs.scandir": "2.1.5",
"fastq": "^1.6.0"
},
"engines": {
"node": ">= 8"
}
},
"node_modules/@pkgr/core": {
"version": "0.2.9",
"resolved": "https://registry.npmjs.org/@pkgr/core/-/core-0.2.9.tgz",
@@ -4370,19 +4332,6 @@
"url": "https://opencollective.com/pkgr"
}
},
"node_modules/@sindresorhus/merge-streams": {
"version": "2.3.0",
"resolved": "https://registry.npmjs.org/@sindresorhus/merge-streams/-/merge-streams-2.3.0.tgz",
"integrity": "sha512-LtoMMhxAlorcGhmFYI+LhPgbPZCkgP6ra1YL604EeF6U98pLlQ3iWIGMdWSC+vWmPBWBNgmDBAhnAobLROJmwg==",
"dev": true,
"license": "MIT",
"engines": {
"node": ">=18"
},
"funding": {
"url": "https://github.com/sponsors/sindresorhus"
}
},
"node_modules/@types/body-parser": {
"version": "1.19.6",
"resolved": "https://registry.npmjs.org/@types/body-parser/-/body-parser-1.19.6.tgz",
@@ -6686,21 +6635,20 @@
"license": "MIT"
},
"node_modules/copy-webpack-plugin": {
"version": "12.0.2",
"resolved": "https://registry.npmjs.org/copy-webpack-plugin/-/copy-webpack-plugin-12.0.2.tgz",
"integrity": "sha512-SNwdBeHyII+rWvee/bTnAYyO8vfVdcSTud4EIb6jcZ8inLeWucJE0DnxXQBjlQ5zlteuuvooGQy3LIyGxhvlOA==",
"version": "14.0.0",
"resolved": "https://registry.npmjs.org/copy-webpack-plugin/-/copy-webpack-plugin-14.0.0.tgz",
"integrity": "sha512-3JLW90aBGeaTLpM7mYQKpnVdgsUZRExY55giiZgLuX/xTQRUs1dOCwbBnWnvY6Q6rfZoXMNwzOQJCSZPppfqXA==",
"dev": true,
"license": "MIT",
"dependencies": {
"fast-glob": "^3.3.2",
"glob-parent": "^6.0.1",
"globby": "^14.0.0",
"normalize-path": "^3.0.0",
"schema-utils": "^4.2.0",
"serialize-javascript": "^6.0.2"
"serialize-javascript": "^7.0.3",
"tinyglobby": "^0.2.12"
},
"engines": {
"node": ">= 18.12.0"
"node": ">= 20.9.0"
},
"funding": {
"type": "opencollective",
@@ -8062,36 +8010,6 @@
"dev": true,
"license": "Apache-2.0"
},
"node_modules/fast-glob": {
"version": "3.3.3",
"resolved": "https://registry.npmjs.org/fast-glob/-/fast-glob-3.3.3.tgz",
"integrity": "sha512-7MptL8U0cqcFdzIzwOTHoilX9x5BrNqye7Z/LuC7kCMRio1EMSyqRK3BEAUD7sXRq4iT4AzTVuZdhgQ2TCvYLg==",
"dev": true,
"license": "MIT",
"dependencies": {
"@nodelib/fs.stat": "^2.0.2",
"@nodelib/fs.walk": "^1.2.3",
"glob-parent": "^5.1.2",
"merge2": "^1.3.0",
"micromatch": "^4.0.8"
},
"engines": {
"node": ">=8.6.0"
}
},
"node_modules/fast-glob/node_modules/glob-parent": {
"version": "5.1.2",
"resolved": "https://registry.npmjs.org/glob-parent/-/glob-parent-5.1.2.tgz",
"integrity": "sha512-AOIgSQCepiJYwP3ARnGx+5VnTu2HBYdzbGP45eLw1vr3zB3vZLeyed1sC9hnbcOc9/SrMyM5RPQrkGz4aS9Zow==",
"dev": true,
"license": "ISC",
"dependencies": {
"is-glob": "^4.0.1"
},
"engines": {
"node": ">= 6"
}
},
"node_modules/fast-json-stable-stringify": {
"version": "2.1.0",
"resolved": "https://registry.npmjs.org/fast-json-stable-stringify/-/fast-json-stable-stringify-2.1.0.tgz",
@@ -8180,16 +8098,6 @@
"node": ">= 4.9.1"
}
},
"node_modules/fastq": {
"version": "1.20.1",
"resolved": "https://registry.npmjs.org/fastq/-/fastq-1.20.1.tgz",
"integrity": "sha512-GGToxJ/w1x32s/D2EKND7kTil4n8OVk/9mycTc4VDza13lOvpUZTGX3mFSCtV9ksdGBVzvsyAVLM6mHFThxXxw==",
"dev": true,
"license": "ISC",
"dependencies": {
"reusify": "^1.0.4"
}
},
"node_modules/faye-websocket": {
"version": "0.11.4",
"resolved": "https://registry.npmjs.org/faye-websocket/-/faye-websocket-0.11.4.tgz",
@@ -8758,37 +8666,6 @@
"url": "https://github.com/sponsors/ljharb"
}
},
"node_modules/globby": {
"version": "14.1.0",
"resolved": "https://registry.npmjs.org/globby/-/globby-14.1.0.tgz",
"integrity": "sha512-0Ia46fDOaT7k4og1PDW4YbodWWr3scS2vAr2lTbsplOt2WkKp0vQbkI9wKis/T5LV/dqPjO3bpS/z6GTJB82LA==",
"dev": true,
"license": "MIT",
"dependencies": {
"@sindresorhus/merge-streams": "^2.1.0",
"fast-glob": "^3.3.3",
"ignore": "^7.0.3",
"path-type": "^6.0.0",
"slash": "^5.1.0",
"unicorn-magic": "^0.3.0"
},
"engines": {
"node": ">=18"
},
"funding": {
"url": "https://github.com/sponsors/sindresorhus"
}
},
"node_modules/globby/node_modules/ignore": {
"version": "7.0.5",
"resolved": "https://registry.npmjs.org/ignore/-/ignore-7.0.5.tgz",
"integrity": "sha512-Hs59xBNfUIunMFgWAbGX5cq6893IbWg4KnrjbYwX3tx0ztorVgTDA6B2sxf8ejHJ4wz8BqGUMYlnzNBer5NvGg==",
"dev": true,
"license": "MIT",
"engines": {
"node": ">= 4"
}
},
"node_modules/gopd": {
"version": "1.2.0",
"resolved": "https://registry.npmjs.org/gopd/-/gopd-1.2.0.tgz",
@@ -11044,16 +10921,6 @@
"dev": true,
"license": "MIT"
},
"node_modules/merge2": {
"version": "1.4.1",
"resolved": "https://registry.npmjs.org/merge2/-/merge2-1.4.1.tgz",
"integrity": "sha512-8q7VEgMJW4J8tcfVPy8g09NcQwZdbwFEqhe/WZkoIzjn/3TGDwtOCYtXGxA3O8tPzpczCCDgv+P2P5y00ZJOOg==",
"dev": true,
"license": "MIT",
"engines": {
"node": ">= 8"
}
},
"node_modules/methods": {
"version": "1.1.2",
"resolved": "https://registry.npmjs.org/methods/-/methods-1.1.2.tgz",
@@ -12638,19 +12505,6 @@
"dev": true,
"license": "MIT"
},
"node_modules/path-type": {
"version": "6.0.0",
"resolved": "https://registry.npmjs.org/path-type/-/path-type-6.0.0.tgz",
"integrity": "sha512-Vj7sf++t5pBD637NSfkxpHSMfWaeig5+DKWLhcqIYx6mWQz5hdJTGDVMQiJcw1ZYkhs7AazKDGpRVji1LJCZUQ==",
"dev": true,
"license": "MIT",
"engines": {
"node": ">=18"
},
"funding": {
"url": "https://github.com/sponsors/sindresorhus"
}
},
"node_modules/pathval": {
"version": "1.1.1",
"resolved": "https://registry.npmjs.org/pathval/-/pathval-1.1.1.tgz",
@@ -12985,37 +12839,6 @@
"url": "https://github.com/sponsors/ljharb"
}
},
"node_modules/queue-microtask": {
"version": "1.2.3",
"resolved": "https://registry.npmjs.org/queue-microtask/-/queue-microtask-1.2.3.tgz",
"integrity": "sha512-NuaNSa6flKT5JaSYQzJok04JzTL1CA6aGhv5rfLW3PgqA+M2ChpZQnAC8h8i4ZFkBS8X5RqkDBHA7r4hej3K9A==",
"dev": true,
"funding": [
{
"type": "github",
"url": "https://github.com/sponsors/feross"
},
{
"type": "patreon",
"url": "https://www.patreon.com/feross"
},
{
"type": "consulting",
"url": "https://feross.org/support"
}
],
"license": "MIT"
},
"node_modules/randombytes": {
"version": "2.1.0",
"resolved": "https://registry.npmjs.org/randombytes/-/randombytes-2.1.0.tgz",
"integrity": "sha512-vYl3iOX+4CKUWuxGi9Ukhie6fsqXqS9FE2Zaic4tNFD2N2QQaXOMFbuKK4QmDHC0JO6B1Zp41J0LpT0oR68amQ==",
"dev": true,
"license": "MIT",
"dependencies": {
"safe-buffer": "^5.1.0"
}
},
"node_modules/range-parser": {
"version": "1.2.1",
"resolved": "https://registry.npmjs.org/range-parser/-/range-parser-1.2.1.tgz",
@@ -13502,17 +13325,6 @@
"node": ">= 4"
}
},
"node_modules/reusify": {
"version": "1.1.0",
"resolved": "https://registry.npmjs.org/reusify/-/reusify-1.1.0.tgz",
"integrity": "sha512-g6QUff04oZpHs0eG5p83rFLhHeV00ug/Yf9nZM6fLeUrPguBTkTQOdpAWWspMh55TZfVQDPaN3NQJfbVRAxdIw==",
"dev": true,
"license": "MIT",
"engines": {
"iojs": ">=1.0.0",
"node": ">=0.10.0"
}
},
"node_modules/rfdc": {
"version": "1.4.1",
"resolved": "https://registry.npmjs.org/rfdc/-/rfdc-1.4.1.tgz",
@@ -13545,30 +13357,6 @@
"node": ">=0.12.0"
}
},
"node_modules/run-parallel": {
"version": "1.2.0",
"resolved": "https://registry.npmjs.org/run-parallel/-/run-parallel-1.2.0.tgz",
"integrity": "sha512-5l4VyZR86LZ/lDxZTR6jqL8AFE2S0IFLMP26AbjsLVADxHdhB/c0GUsH+y39UfCi3dzz8OlQuPmnaJOMoDHQBA==",
"dev": true,
"funding": [
{
"type": "github",
"url": "https://github.com/sponsors/feross"
},
{
"type": "patreon",
"url": "https://www.patreon.com/feross"
},
{
"type": "consulting",
"url": "https://feross.org/support"
}
],
"license": "MIT",
"dependencies": {
"queue-microtask": "^1.2.2"
}
},
"node_modules/rxjs": {
"version": "7.8.2",
"resolved": "https://registry.npmjs.org/rxjs/-/rxjs-7.8.2.tgz",
@@ -13842,13 +13630,13 @@
}
},
"node_modules/serialize-javascript": {
"version": "6.0.2",
"resolved": "https://registry.npmjs.org/serialize-javascript/-/serialize-javascript-6.0.2.tgz",
"integrity": "sha512-Saa1xPByTTq2gdeFZYLLo+RFE35NHZkAbqZeWNd3BpzppeVisAqpDjcp8dyf6uIvEqJRd46jemmyA4iFIeVk8g==",
"version": "7.0.5",
"resolved": "https://registry.npmjs.org/serialize-javascript/-/serialize-javascript-7.0.5.tgz",
"integrity": "sha512-F4LcB0UqUl1zErq+1nYEEzSHJnIwb3AF2XWB94b+afhrekOUijwooAYqFyRbjYkm2PAKBabx6oYv/xDxNi8IBw==",
"dev": true,
"license": "BSD-3-Clause",
"dependencies": {
"randombytes": "^2.1.0"
"engines": {
"node": ">=20.0.0"
}
},
"node_modules/serve-index": {
@@ -14260,19 +14048,6 @@
"simple-concat": "^1.0.0"
}
},
"node_modules/slash": {
"version": "5.1.0",
"resolved": "https://registry.npmjs.org/slash/-/slash-5.1.0.tgz",
"integrity": "sha512-ZA6oR3T/pEyuqwMgAKT0/hAv8oAXckzbkmR0UkUosQ+Mc4RxGoJkRmwHgHufaenlyAgE1Mxgpdcrf75y6XcnDg==",
"dev": true,
"license": "MIT",
"engines": {
"node": ">=14.16"
},
"funding": {
"url": "https://github.com/sponsors/sindresorhus"
}
},
"node_modules/sockjs": {
"version": "0.3.24",
"resolved": "https://registry.npmjs.org/sockjs/-/sockjs-0.3.24.tgz",
@@ -15311,19 +15086,6 @@
"node": ">=4"
}
},
"node_modules/unicorn-magic": {
"version": "0.3.0",
"resolved": "https://registry.npmjs.org/unicorn-magic/-/unicorn-magic-0.3.0.tgz",
"integrity": "sha512-+QBBXBCvifc56fsbuxZQ6Sic3wqqc3WWaqxs58gvJrcOuN83HGTCwz3oS5phzU9LthRNE9VrJCFCLUgHeeFnfA==",
"dev": true,
"license": "MIT",
"engines": {
"node": ">=18"
},
"funding": {
"url": "https://github.com/sponsors/sindresorhus"
}
},
"node_modules/universalify": {
"version": "2.0.1",
"resolved": "https://registry.npmjs.org/universalify/-/universalify-2.0.1.tgz",
+1 -1
View File
@@ -36,7 +36,7 @@
"@types/office-runtime": "^1.0.35",
"acorn": "^8.11.3",
"babel-loader": "^9.1.3",
"copy-webpack-plugin": "^12.0.2",
"copy-webpack-plugin": "^14.0.0",
"eslint-plugin-office-addins": "^4.0.3",
"file-loader": "^6.2.0",
"html-loader": "^5.0.0",
+1 -1
View File
@@ -49,7 +49,7 @@ RUN --mount=type=cache,target=/root/.cache/uv \
ENV PATH="/app/.venv/bin:$PATH"
CMD ["python", "metadata_collector.py", "dev"]
CMD ["python", "multi_user_transcriber.py", "dev"]
# ---- Production image ----
-5
View File
@@ -25,8 +25,6 @@ from livekit.agents import (
)
from livekit.plugins import deepgram, silero
import voxtral_vllm_stt
load_dotenv()
logger = logging.getLogger("transcriber")
@@ -48,9 +46,6 @@ def create_stt_provider():
)
elif STT_PROVIDER == "kyutai":
_stt_instance = kyutai.STT(base_url=os.getenv("KYUTAI_STT_BASE_URL"))
elif STT_PROVIDER == "voxtral-vllm":
# The plugin resolves base_url / model / api_key from the environment.
_stt_instance = voxtral_vllm_stt.STT()
else:
raise ValueError(f"Unknown STT_PROVIDER: {STT_PROVIDER}")
+1 -1
View File
@@ -1,7 +1,7 @@
[project]
name = "agents"
version = "1.15.0"
version = "1.16.0"
requires-python = ">=3.12"
dependencies = [
"livekit-agents==1.4.5",
+15 -15
View File
@@ -9,7 +9,7 @@ resolution-markers = [
[[package]]
name = "agents"
version = "1.15.0"
version = "1.16.0"
source = { virtual = "." }
dependencies = [
{ name = "livekit-agents" },
@@ -28,7 +28,7 @@ dev = [
[package.metadata]
requires-dist = [
{ name = "livekit-agents", specifier = ">=1.4.5" },
{ name = "livekit-agents", specifier = "==1.4.5" },
{ name = "livekit-plugins-deepgram", specifier = "==1.4.5" },
{ name = "livekit-plugins-kyutai-lasuite", specifier = "==0.0.6" },
{ name = "livekit-plugins-silero", specifier = "==1.4.5" },
@@ -731,7 +731,7 @@ wheels = [
[[package]]
name = "livekit"
version = "1.1.7"
version = "1.1.2"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "aiofiles" },
@@ -739,18 +739,18 @@ dependencies = [
{ name = "protobuf" },
{ name = "types-protobuf" },
]
sdist = { url = "https://files.pythonhosted.org/packages/e1/1e/3ad724abacb6514fc64c7b4a8c82a41780cbc82baee621b8dcc92dded005/livekit-1.1.7.tar.gz", hash = "sha256:29e5f13e2639c041f18e29d1066019a6380031391b6363509f06e3dbef8426a6", size = 332704, upload-time = "2026-04-27T13:49:48.432Z" }
sdist = { url = "https://files.pythonhosted.org/packages/9c/9c/0a9b9e1f88226df372b9ccc58868ce9a1eb97c8bf5257463a643dda2a6da/livekit-1.1.2.tar.gz", hash = "sha256:ef826e94dd039767fcabc2f0d810b1b2335c9cf249f52320b6ab018b06d5ccd7", size = 320006, upload-time = "2026-02-17T01:18:46.828Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/fd/f8/176fe3d5569ed03b2c60991f4a90edfd2421a5ad0c4a9a70974d409b3329/livekit-1.1.7-py3-none-macosx_10_15_x86_64.whl", hash = "sha256:26df0760e9b9d3eeb0d773614bc344b6aae7d37c60a9b0c93b1c58dcb6bb2f6b", size = 10127354, upload-time = "2026-04-27T13:49:35.805Z" },
{ url = "https://files.pythonhosted.org/packages/29/f2/05bd6f3729ecb564ddc9052afc43c3b9e39a43d24f7821946bc78a5dddd2/livekit-1.1.7-py3-none-macosx_11_0_arm64.whl", hash = "sha256:e49e79d39b2c74104c377b6e3cb789d47dd92cb993327234966c41f19df28292", size = 8954208, upload-time = "2026-04-27T13:49:38.359Z" },
{ url = "https://files.pythonhosted.org/packages/58/32/99f52a9c71dcc3b3370783ad427e5974b9491948b6d9beb5b3f933df8f1e/livekit-1.1.7-py3-none-manylinux_2_28_aarch64.whl", hash = "sha256:c3db851f510edac945b9bc402f6629d5969b4f3252e8d0b24dd158d9e9a5485d", size = 9962579, upload-time = "2026-04-27T13:49:40.28Z" },
{ url = "https://files.pythonhosted.org/packages/bf/9e/53768dd73f1ce2d3eff7d6c1d87c3531f795c57880a3b51eb5cee1aec6fa/livekit-1.1.7-py3-none-manylinux_2_28_x86_64.whl", hash = "sha256:6c227e2f840aa547224e2fbd45f953c8a0b30939a18e1d56deca8c4268c95804", size = 11355003, upload-time = "2026-04-27T13:49:42.881Z" },
{ url = "https://files.pythonhosted.org/packages/42/55/2c68732dbcdb3fc1a58592db27697c6d27e2f77ade7eaeb818e88637020b/livekit-1.1.7-py3-none-win_amd64.whl", hash = "sha256:3b0bcc48430cb2138eade01f6b59ddd3be1c1e90f6c251291d0dd4dc1623fc0d", size = 10696832, upload-time = "2026-04-27T13:49:46.362Z" },
{ url = "https://files.pythonhosted.org/packages/07/7b/1b2d448d8976b14794bfba3d003e1451c79afa77e6b9fa1593f19175ef90/livekit-1.1.2-py3-none-macosx_10_15_x86_64.whl", hash = "sha256:78be23f3f6315354aaacee664eb19b793009bc06faa8184ad9c07cffbe8d7f74", size = 9844157, upload-time = "2026-02-17T01:18:34.036Z" },
{ url = "https://files.pythonhosted.org/packages/f9/fb/19f7fc4a7df3b1385ba2e32b907c5a502fe01c10e6ee2fb76629c068667d/livekit-1.1.2-py3-none-macosx_11_0_arm64.whl", hash = "sha256:656f4d9e4692f3d7ab2e51b0bbf4ec03b356a487b7ff220576dab496e60f99f3", size = 8651703, upload-time = "2026-02-17T01:18:36.416Z" },
{ url = "https://files.pythonhosted.org/packages/29/59/ac6a3987bfd11687f86156627a8c7f6a0047a72877748facbd427e63b157/livekit-1.1.2-py3-none-manylinux_2_28_aarch64.whl", hash = "sha256:a20e681d8a4929e27df69f818888ae649700c9d52a262abbec296c84937bc337", size = 14085891, upload-time = "2026-02-17T01:18:38.561Z" },
{ url = "https://files.pythonhosted.org/packages/0a/a5/680548ef7bf5034144ae01f1f742a6c49e4428942b3119ac6553207fea9b/livekit-1.1.2-py3-none-manylinux_2_28_x86_64.whl", hash = "sha256:0e4d8105a0c317b513a118b48340b5773239103cb6305768451ef99ac7567823", size = 11448902, upload-time = "2026-02-17T01:18:42.006Z" },
{ url = "https://files.pythonhosted.org/packages/00/05/e484f8fc079c5de7690e58ed48f44e77436b8732c3050da742bdfca51a5c/livekit-1.1.2-py3-none-win_amd64.whl", hash = "sha256:dd4a436fa16de589353bfbabde91068ab64241afd05b04f21fb1f22bfe155dc0", size = 10394562, upload-time = "2026-02-17T01:18:44.589Z" },
]
[[package]]
name = "livekit-agents"
version = "1.5.8"
version = "1.4.5"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "aiofiles" },
@@ -782,9 +782,9 @@ dependencies = [
{ name = "typing-extensions" },
{ name = "watchfiles" },
]
sdist = { url = "https://files.pythonhosted.org/packages/bb/23/8af550199aef211304290596c1ce073de9ac22d6b2beea51ca362d70b2eb/livekit_agents-1.5.8.tar.gz", hash = "sha256:ab8951fb7a4f79f81c13f962a401a76c9ecbcb79c9ebd5cad5f0554062b70b6d", size = 2494457, upload-time = "2026-05-05T18:46:42.965Z" }
sdist = { url = "https://files.pythonhosted.org/packages/9c/06/1a23902ffea2742e60e30f957f3a96766be125892a2573ea1754051b3d8d/livekit_agents-1.4.5.tar.gz", hash = "sha256:19fdf45a2c4d02d86b4ad132b73bf3f58eadb75f79819e5015e6fadfb1263852", size = 2416100, upload-time = "2026-03-11T06:45:03.912Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/37/1e/3f5e2580f9ffdb5267a7c9be003f8f711ea58c4e0764a74999c800ed3d0e/livekit_agents-1.5.8-py3-none-any.whl", hash = "sha256:737a2d517cf2d6c1d22221d720ed12fcd47a21019abb4b5b20533407b57bb9ea", size = 2594846, upload-time = "2026-05-05T18:46:40.156Z" },
{ url = "https://files.pythonhosted.org/packages/71/ad/a7613ddd5cfb29e015913ae44b804505d41c1d02d19ffeaf2f7015d23664/livekit_agents-1.4.5-py3-none-any.whl", hash = "sha256:aff0b61dbd24b1a100cccca8e2a0056f414b861ce71ba8e4338bc5f89e149f7a", size = 2499369, upload-time = "2026-03-11T06:45:01.772Z" },
]
[package.optional-dependencies]
@@ -878,15 +878,15 @@ wheels = [
[[package]]
name = "livekit-protocol"
version = "1.1.8"
version = "1.1.6"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "protobuf" },
{ name = "types-protobuf" },
]
sdist = { url = "https://files.pythonhosted.org/packages/6b/17/761372e1db01925bda77c679ffde19a9d128a4ba53d864dc50a0a9f93ce8/livekit_protocol-1.1.8.tar.gz", hash = "sha256:b3850c09847d54aaa538c87a90ffb5cf267123a0725f2b03874e39e88fc59ccb", size = 98588, upload-time = "2026-05-05T09:12:33.171Z" }
sdist = { url = "https://files.pythonhosted.org/packages/5b/28/8a7ee5a4126476a0ce770a482274d51fc81ae0d9f578c07f07d6655acfc4/livekit_protocol-1.1.6.tar.gz", hash = "sha256:43648863440a6ce064f7f8b8b287dda8f5ac82b8da64738b20f42cad51e70f9b", size = 93925, upload-time = "2026-04-20T15:48:43.032Z" }
wheels = [
{ url = "https://files.pythonhosted.org/packages/6d/54/fca19cd45d0c74bb5ef88b0257cf594ec5549192a15957dd22596acdec32/livekit_protocol-1.1.8-py3-none-any.whl", hash = "sha256:dc379b3e713dc8d8642f694c72d9a38b41220f05b2ce8fc6662f445d20953c17", size = 121119, upload-time = "2026-05-05T09:12:31.787Z" },
{ url = "https://files.pythonhosted.org/packages/76/36/525cd9364036617aa97c597c93d7bddb6be311b86288cd54de77e4892544/livekit_protocol-1.1.6-py3-none-any.whl", hash = "sha256:286ac0bd555b6b75cd8a0c5d417572d7dab524129fcea43080202d170f957710", size = 115012, upload-time = "2026-04-20T15:48:41.312Z" },
]
[[package]]
-474
View File
@@ -1,474 +0,0 @@
"""LiveKit STT plugin for Voxtral Realtime served via vLLM (/v1/realtime).
vLLM exposes Voxtral Realtime over a WebSocket that follows the OpenAI Realtime
API protocol (not Mistral's proprietary realtime protocol).
"""
from __future__ import annotations
import asyncio
import base64
import json
import logging
import os
import weakref
from collections import deque
from dataclasses import dataclass, field
import websockets
from livekit.agents import (
DEFAULT_API_CONNECT_OPTIONS,
APIConnectionError,
APIConnectOptions,
APIStatusError,
stt,
utils,
)
from livekit.agents import (
vad as vad_module,
)
from livekit.agents.types import NOT_GIVEN, NotGivenOr
from livekit.agents.utils import is_given
logger = logging.getLogger("voxtral-vllm-stt")
SAMPLE_RATE = 16000
NUM_CHANNELS = 1
CHUNK_SAMPLES = 1600 # 100 ms @ 16 kHz mono
PREROLL_CHUNKS = 5 # keep 500 ms of audio before start of speech as detected by VAD
# Reconnect policy: exponential backoff capped at MAX, give up after MAX_ATTEMPTS
# consecutive failures (a successful handshake resets the counter).
RECONNECT_BACKOFF_BASE_S = 0.5
RECONNECT_BACKOFF_MAX_S = 8.0
RECONNECT_MAX_ATTEMPTS = 5
@dataclass
class _STTOptions:
base_url: str
model: str
api_key: str | None
target_streaming_delay_ms: int | None
@dataclass
class _PendingUtterance:
"""An utterance in flight on the shared websocket used for reconnect.
`sent_chunks` holds every chunk we have already enqueued for send on this
or a prior connection; on reconnect we replay them before resuming reads
from `queue`. vLLM concatenates `input_audio_buffer.append` events into a
single audio buffer per generation, so duplicates from a partial prior send
are harmless.
"""
queue: asyncio.Queue[bytes | None]
sent_chunks: list[bytes] = field(default_factory=list)
ended: bool = False
class STT(stt.STT):
"""LiveKit STT speaking the OpenAI Realtime protocol served by vLLM."""
def __init__(
self,
*,
base_url: NotGivenOr[str] = NOT_GIVEN,
model: NotGivenOr[str] = NOT_GIVEN,
api_key: NotGivenOr[str] = NOT_GIVEN,
target_streaming_delay_ms: NotGivenOr[int] = NOT_GIVEN,
vad: vad_module.VAD | None = None,
) -> None:
"""Build the STT.
Args:
base_url: WebSocket URL of the vLLM realtime endpoint, e.g.
ws://example:8000/v1/realtime. Falls back to $VOXTRAL_VLLM_BASE_URL.
model: Model name exposed by vLLM, default
mistralai/Voxtral-Mini-4B-Realtime-2602.
api_key: Optional bearer token. Falls back to $VOXTRAL_VLLM_API_KEY.
target_streaming_delay_ms: Target streaming delay in ms forwarded to
vLLM via session.update. Falls back to
$VOXTRAL_VLLM_TARGET_STREAMING_DELAY_MS, else server default.
vad: Voice Activity Detector. If omitted, Silero VAD is loaded.
"""
super().__init__(
capabilities=stt.STTCapabilities(streaming=True, interim_results=True)
)
resolved_url = (
base_url
if is_given(base_url)
else os.environ.get(
"VOXTRAL_VLLM_BASE_URL", "ws://127.0.0.1:8000/v1/realtime"
)
)
resolved_model = (
model
if is_given(model)
else os.environ.get(
"VOXTRAL_VLLM_MODEL", "mistralai/Voxtral-Mini-4B-Realtime-2602"
)
)
resolved_key = (
api_key if is_given(api_key) else os.environ.get("VOXTRAL_VLLM_API_KEY")
)
resolved_delay = (
target_streaming_delay_ms
if is_given(target_streaming_delay_ms)
else (
int(os.environ["VOXTRAL_VLLM_TARGET_STREAMING_DELAY_MS"])
if os.environ.get("VOXTRAL_VLLM_TARGET_STREAMING_DELAY_MS")
else None
)
)
if vad is None:
try:
from livekit.plugins.silero import VAD as SileroVAD # noqa: PLC0415
except ImportError as exc:
raise ImportError(
"livekit-plugins-silero is required for vLLM Voxtral realtime "
"(no server-side endpointing)."
) from exc
vad = SileroVAD.load()
self._vad = vad
self._opts = _STTOptions(
base_url=resolved_url,
model=resolved_model,
api_key=resolved_key,
target_streaming_delay_ms=resolved_delay,
)
self._streams: weakref.WeakSet[SpeechStream] = weakref.WeakSet()
@property
def model(self) -> str:
"""Return the configured vLLM model name."""
return self._opts.model
@property
def provider(self) -> str:
"""Return the provider identifier."""
return "vllm-voxtral-realtime"
async def _recognize_impl(self, *_args, **_kwargs) -> stt.SpeechEvent:
raise NotImplementedError(
"vLLM Voxtral Realtime STT only supports streaming recognition."
)
def stream(
self,
*,
conn_options: APIConnectOptions = DEFAULT_API_CONNECT_OPTIONS,
) -> SpeechStream:
"""Open a new streaming recognition stream."""
s = SpeechStream(
stt=self,
opts=self._opts,
vad_instance=self._vad,
conn_options=conn_options,
)
self._streams.add(s)
return s
class SpeechStream(stt.RecognizeStream):
def __init__(
self,
*,
stt: STT,
opts: _STTOptions,
vad_instance: vad_module.VAD,
conn_options: APIConnectOptions,
) -> None:
"""Init the speech stream."""
super().__init__(stt=stt, conn_options=conn_options, sample_rate=SAMPLE_RATE)
self._opts = opts
self._vad = vad_instance
self._utterance_q: asyncio.Queue[bytes | None] | None = None
self._speaking = False
self._preroll: deque[bytes] = deque(maxlen=PREROLL_CHUNKS)
# Voxtral realtime is strictly sequential: only one generation runs at a
# time, and a new `commit` is ignored while the previous one is still
# producing. We queue per-utterance audio buffers here and let the
# pipeline process them one by one on the shared websocket.
self._utterance_chan: asyncio.Queue[asyncio.Queue[bytes | None] | None] = (
asyncio.Queue()
)
@utils.log_exceptions(logger=logger)
async def _run(self) -> None:
vad_stream = self._vad.stream()
bstream = utils.audio.AudioByteStream(
sample_rate=SAMPLE_RATE,
num_channels=NUM_CHANNELS,
samples_per_channel=CHUNK_SAMPLES,
)
async def input_task() -> None:
async for data in self._input_ch:
if isinstance(data, self._FlushSentinel):
for frame in bstream.flush():
self._handle_chunk(frame.data.tobytes())
continue
vad_stream.push_frame(data)
for frame in bstream.write(data.data.tobytes()):
self._handle_chunk(frame.data.tobytes())
vad_stream.end_input()
async def vad_task() -> None:
async for ev in vad_stream:
if ev.type == vad_module.VADEventType.START_OF_SPEECH:
self._on_start_of_speech()
elif ev.type == vad_module.VADEventType.END_OF_SPEECH:
self._on_end_of_speech()
pipeline_t = asyncio.create_task(self._utterance_pipeline())
try:
await asyncio.gather(input_task(), vad_task())
# signal end-of-stream; pipeline finishes pending utterances first
self._utterance_chan.put_nowait(None)
await pipeline_t
except (APIStatusError, APIConnectionError, asyncio.CancelledError):
raise
except Exception as exc:
logger.exception("vLLM realtime stream failed")
raise APIConnectionError() from exc
finally:
if not pipeline_t.done():
pipeline_t.cancel()
try:
await pipeline_t
except asyncio.CancelledError:
# CancelledError is the expected flow on cancel()
pass
except Exception:
logger.exception("utterance pipeline failed during finalize")
await vad_stream.aclose()
def _handle_chunk(self, chunk: bytes) -> None:
self._preroll.append(chunk)
if self._speaking and self._utterance_q is not None:
self._utterance_q.put_nowait(chunk)
def _on_start_of_speech(self) -> None:
if self._speaking:
return
self._speaking = True
q: asyncio.Queue[bytes | None] = asyncio.Queue()
for chunk in self._preroll:
q.put_nowait(chunk)
self._utterance_q = q
self._utterance_chan.put_nowait(q)
self._event_ch.send_nowait(
stt.SpeechEvent(type=stt.SpeechEventType.START_OF_SPEECH)
)
def _on_end_of_speech(self) -> None:
if not self._speaking:
return
self._speaking = False
if self._utterance_q is not None:
self._utterance_q.put_nowait(None)
self._utterance_q = None
self._event_ch.send_nowait(
stt.SpeechEvent(type=stt.SpeechEventType.END_OF_SPEECH)
)
async def _handshake(self, ws: websockets.ClientConnection) -> str:
created = json.loads(await ws.recv())
if created.get("type") != "session.created":
raise APIStatusError(
f"expected session.created, got {created}",
status_code=500,
body=created,
)
session_update: dict = {"type": "session.update", "model": self._opts.model}
if self._opts.target_streaming_delay_ms is not None:
session_update["target_streaming_delay_ms"] = (
self._opts.target_streaming_delay_ms
)
await ws.send(json.dumps(session_update))
return created.get("id", "")
def _auth_headers(self) -> dict[str, str]:
if self._opts.api_key:
return {"Authorization": f"Bearer {self._opts.api_key}"}
return {}
async def _utterance_pipeline(self) -> None:
# Owns the websocket lifecycle. On drop, reopens and resumes the
# in-flight utterance (if any) by replaying its already-sent chunks.
pending: _PendingUtterance | None = None
attempt = 0
while True:
try:
async with websockets.connect(
self._opts.base_url,
additional_headers=self._auth_headers(),
open_timeout=self._conn_options.timeout,
) as ws:
request_id = await self._handshake(ws)
attempt = 0
while True:
if pending is None:
q = await self._utterance_chan.get()
if q is None:
return
pending = _PendingUtterance(queue=q)
await self._process_utterance(ws, pending, request_id)
pending = None
except (websockets.WebSocketException, OSError, TimeoutError) as exc:
attempt += 1
if attempt > RECONNECT_MAX_ATTEMPTS:
logger.exception(
"vLLM realtime: giving up after %d reconnect attempts",
RECONNECT_MAX_ATTEMPTS,
)
raise APIConnectionError() from exc
backoff = min(
RECONNECT_BACKOFF_BASE_S * (2 ** (attempt - 1)),
RECONNECT_BACKOFF_MAX_S,
)
if pending is None:
logger.warning(
"vLLM WS connection lost between utterances "
"(attempt %d/%d): %s; retrying in %.1fs",
attempt,
RECONNECT_MAX_ATTEMPTS,
exc,
backoff,
)
else:
logger.warning(
"vLLM WS dropped mid-utterance (%d chunks buffered, "
"ended=%s, attempt %d/%d): %s; retrying in %.1fs",
len(pending.sent_chunks),
pending.ended,
attempt,
RECONNECT_MAX_ATTEMPTS,
exc,
backoff,
)
await asyncio.sleep(backoff)
async def _process_utterance(
self,
ws: websockets.ClientConnection,
pending: _PendingUtterance,
request_id: str,
) -> None:
# Start a fresh generation. Safe to send here: the previous utterance's
# transcription.done has already been received (we await it below), so
# the server-side generation_task is done and won't ignore this commit.
await ws.send(json.dumps({"type": "input_audio_buffer.commit"}))
send_t = asyncio.create_task(self._send_audio(ws, pending))
try:
await self._receive_one_transcription(ws, request_id)
finally:
if not send_t.done():
send_t.cancel()
try:
await send_t
except (asyncio.CancelledError, websockets.WebSocketException):
pass
except Exception:
logger.exception("send-audio task failed during finalize")
@staticmethod
async def _send_audio(
ws: websockets.ClientConnection, pending: _PendingUtterance
) -> None:
# Replay anything already sent on a previous (now-dead) connection.
# sent_chunks is appended before send, so a chunk that failed to send
# last time is still present and gets retried here.
for chunk in pending.sent_chunks:
await ws.send(
json.dumps(
{
"type": "input_audio_buffer.append",
"audio": base64.b64encode(chunk).decode("ascii"),
}
)
)
if pending.ended:
await ws.send(
json.dumps({"type": "input_audio_buffer.commit", "final": True})
)
return
while True:
chunk = await pending.queue.get()
if chunk is None:
pending.ended = True
await ws.send(
json.dumps({"type": "input_audio_buffer.commit", "final": True})
)
return
pending.sent_chunks.append(chunk)
await ws.send(
json.dumps(
{
"type": "input_audio_buffer.append",
"audio": base64.b64encode(chunk).decode("ascii"),
}
)
)
async def _receive_one_transcription(
self, ws: websockets.ClientConnection, request_id: str
) -> None:
# Use recv() rather than `async for`: the latter swallows
# ConnectionClosed on close-mid-iteration, which would let a dropped
# WS look like a clean "no transcription" return.
current_text = ""
while True:
raw = await ws.recv()
data = json.loads(raw)
event_type = data.get("type")
if event_type == "transcription.delta":
delta = data.get("delta", "")
if not delta:
continue
current_text += delta
self._event_ch.send_nowait(
stt.SpeechEvent(
type=stt.SpeechEventType.INTERIM_TRANSCRIPT,
request_id=request_id,
alternatives=[stt.SpeechData(text=current_text, language="")],
)
)
elif event_type == "transcription.done":
final_text = data.get("text") or current_text
self._event_ch.send_nowait(
stt.SpeechEvent(
type=stt.SpeechEventType.FINAL_TRANSCRIPT,
request_id=request_id,
alternatives=[stt.SpeechData(text=final_text, language="")],
)
)
usage = data.get("usage") or {}
self._event_ch.send_nowait(
stt.SpeechEvent(
type=stt.SpeechEventType.RECOGNITION_USAGE,
request_id=request_id,
recognition_usage=stt.RecognitionUsage(
audio_duration=float(
usage.get("audio_seconds")
or usage.get("prompt_audio_seconds")
or 0
),
input_tokens=int(usage.get("prompt_tokens") or 0),
output_tokens=int(usage.get("completion_tokens") or 0),
),
)
)
return
elif event_type == "error":
err = data.get("error")
raise APIStatusError(str(err), status_code=500, body=data)
+30
View File
@@ -136,3 +136,33 @@ class FilePermission(IsAuthenticated):
raise Http404
return obj.get_abilities(request.user).get(view.action, False)
class CanMuteParticipant(permissions.BasePermission):
"""
Grant muting rights based on role or room configuration.
- Admins and owners can always mute.
- When `everyone_can_mute` is enabled on the room, any participant
currently in the room (proven by a valid LiveKit token for that room)
can mute.
"""
def has_object_permission(self, request, view, obj):
"""Check if the requesting user is allowed to mute a participant in the given room."""
is_livekit_token_auth = request.auth and hasattr(request.auth, "video")
# Always allow admins/owners when authenticated with session cookie
if not is_livekit_token_auth and obj.is_administrator_or_owner(request.user):
return True
everyone_can_mute = obj.configuration.get("everyone_can_mute", True)
if not everyone_can_mute:
return False
if not is_livekit_token_auth:
return False
# LiveKit token scoped to this room
return request.auth.video.room == str(obj.id)
+9 -14
View File
@@ -13,7 +13,7 @@ from django.core.exceptions import SuspiciousOperation
from django.utils.translation import gettext_lazy as _
from django_pydantic_field.rest_framework import SchemaField
from pydantic import BaseModel, Field
from pydantic import BaseModel, Field, field_serializer
from pydantic import ValidationError as PydanticValidationError
from rest_framework import serializers
from rest_framework.exceptions import PermissionDenied
@@ -166,11 +166,6 @@ class RoomSerializer(serializers.ModelSerializer):
)
output["accesses"] = access_serializer.data
configuration = output["configuration"]
if not is_admin_or_owner:
del output["configuration"]
should_access_room = (
(
instance.access_level == models.RoomAccessLevel.TRUSTED
@@ -187,7 +182,7 @@ class RoomSerializer(serializers.ModelSerializer):
room_id=room_id,
user=request.user,
username=username,
configuration=configuration,
configuration=output["configuration"],
is_admin_or_owner=is_admin_or_owner,
)
else:
@@ -317,9 +312,7 @@ class MuteParticipantSerializer(BaseParticipantsManagementSerializer):
)
RoomConfigurationTrackSource = Literal[
"camera", "microphone", "screen_share", "screen_share_audio"
]
TrackSource = Literal["camera", "microphone", "screen_share", "screen_share_audio"]
class RoomConfiguration(BaseModel):
@@ -328,14 +321,12 @@ class RoomConfiguration(BaseModel):
Unknown fields are rejected.
"""
can_publish_sources: list[RoomConfigurationTrackSource] | None = None
can_publish_sources: list[TrackSource] | None = None
everyone_can_mute: bool | None = None
model_config = {"extra": "forbid"}
TrackSource = Literal["SCREEN_SHARE", "SCREEN_SHARE_AUDIO", "CAMERA", "MICROPHONE"]
class ParticipantPermission(BaseModel):
"""Mirror the LiveKit ParticipantPermission protobuf.
@@ -355,6 +346,10 @@ class ParticipantPermission(BaseModel):
model_config = {"extra": "forbid"}
@field_serializer("can_publish_sources")
def _serialize_sources(self, sources: list[str]) -> list[str]:
return [s.upper() for s in sources]
class UpdateParticipantSerializer(BaseParticipantsManagementSerializer):
"""Validate participant update data."""
+66 -1
View File
@@ -33,6 +33,7 @@ from rest_framework import (
from rest_framework import (
status as drf_status,
)
from rest_framework.settings import api_settings
from core import enums, models, utils
from core.api.filters import ListFileFilter
@@ -76,6 +77,11 @@ from core.services.participants_management import (
ParticipantsManagementException,
)
from core.services.room_creation import RoomCreation
from core.services.room_management import (
RoomManagement,
RoomManagementException,
RoomNotFoundException,
)
from core.services.subtitle import SubtitleException, SubtitleService
from core.tasks.file import process_file_deletion
@@ -299,6 +305,41 @@ class RoomViewSet(
if callback_id := self.request.data.get("callback_id"):
RoomCreation().persist_callback_state(callback_id, room)
def perform_update(self, serializer):
"""Persist the room update, then sync metadata to LiveKit."""
old_configuration = serializer.instance.configuration
old_access_level = serializer.instance.access_level
room = serializer.save()
if (
room.configuration == old_configuration
and room.access_level == old_access_level
):
return
metadata = {
"configuration": room.configuration,
"access_level": room.access_level,
}
try:
RoomManagement().update_metadata(
room_name=str(room.id),
metadata=metadata,
)
except RoomNotFoundException:
logger.info(
"LiveKit room %s does not exist yet, skipping metadata sync",
room.id,
)
except RoomManagementException:
logger.warning(
"Failed to sync metadata to LiveKit for room %s",
room.id,
)
@decorators.action(
detail=True,
methods=["post"],
@@ -614,7 +655,11 @@ class RoomViewSet(
methods=["post"],
url_path="mute-participant",
url_name="mute-participant",
permission_classes=[permissions.HasPrivilegesOnRoom],
permission_classes=[permissions.CanMuteParticipant],
authentication_classes=[
LiveKitTokenAuthentication,
*api_settings.DEFAULT_AUTHENTICATION_CLASSES,
],
)
def mute_participant(self, request, pk=None): # pylint: disable=unused-argument
"""Mute a specific track for a participant in the room."""
@@ -623,6 +668,26 @@ class RoomViewSet(
serializer = serializers.MuteParticipantSerializer(data=request.data)
serializer.is_valid(raise_exception=True)
# TEMPORARY: a LiveKit token proves access was granted, not that the caller
# joined. Cross-check identity against the live participant list until auth
# is hardened. Skipped for non-LiveKit auth backends.
caller_identity = getattr(request.auth, "identity", None)
if caller_identity is not None:
try:
ParticipantsManagement().check_if_in_meeting(
room_name=str(room.pk),
identity=caller_identity,
)
except (ParticipantNotFoundException, ParticipantsManagementException):
logger.warning(
"Failed to verify caller presence for mute in room %s; denying",
room.pk,
)
return drf_response.Response(
{"error": "Could not verify caller presence"},
status=drf_status.HTTP_403_FORBIDDEN,
)
try:
ParticipantsManagement().mute(
room_name=str(room.pk),
+1
View File
@@ -388,6 +388,7 @@ class Room(Resource):
choices=RoomAccessLevel.choices,
default=settings.RESOURCE_DEFAULT_ACCESS_LEVEL,
)
# Public configuration exposed to any room participant via the API
configuration = models.JSONField(
blank=True,
default=dict,
@@ -15,6 +15,7 @@ from livekit.api import (
TwirpError,
UpdateParticipantRequest,
)
from livekit.protocol.models import ParticipantInfo
from core import utils
@@ -154,3 +155,44 @@ class ParticipantsManagement:
finally:
await lkapi.aclose()
@async_to_sync
async def check_if_in_meeting(self, room_name: str, identity: str) -> bool:
"""Check whether `identity` is currently a participant in `room_name`.
Raises ParticipantsManagementException for unexpected LiveKit errors
so callers can fail closed rather than silently allowing the action.
"""
if not room_name or not identity:
return False
lkapi = utils.create_livekit_client()
try:
participant = await lkapi.room.get_participant(
RoomParticipantIdentity(
room=room_name,
identity=identity,
)
)
except TwirpError as e:
if e.code == "not_found":
raise ParticipantNotFoundException("Participant does not exist") from e
logger.exception(
"Unexpected error checking participant %s in room %s",
identity,
room_name,
)
raise ParticipantsManagementException(
"Could not verify participant presence"
) from e
finally:
await lkapi.aclose()
return (
participant is not None
and participant.state != ParticipantInfo.State.DISCONNECTED
)
@@ -0,0 +1,64 @@
"""Room management service for LiveKit rooms."""
# pylint: disable=no-name-in-module
import json
from logging import getLogger
from typing import Dict, Optional
from asgiref.sync import async_to_sync
from livekit.api import (
TwirpError,
UpdateRoomMetadataRequest,
)
from core import utils
logger = getLogger(__name__)
class RoomManagementException(Exception):
"""Exception raised when a room management operation fails."""
class RoomNotFoundException(RoomManagementException):
"""Raised when the target room does not exist in LiveKit."""
class RoomManagement:
"""Service for managing LiveKit rooms."""
@async_to_sync
async def update_metadata(self, room_name: str, metadata: Optional[Dict] = None):
"""Update a LiveKit room's metadata.
The `room_name` corresponds to the LiveKit room identifier
(i.e. the Room model's UUID as a string).
"""
lkapi = utils.create_livekit_client()
try:
await lkapi.room.update_room_metadata(
UpdateRoomMetadataRequest(
room=room_name,
metadata=json.dumps(metadata) if metadata is not None else "",
)
)
except TwirpError as e:
if e.code == "not_found":
logger.warning(
"Room %s not found in LiveKit, skipping metadata update",
room_name,
)
raise RoomNotFoundException("Room does not exist") from e
logger.exception(
"Unexpected error updating metadata for room %s",
room_name,
)
raise RoomManagementException("Could not update room metadata") from e
finally:
await lkapi.aclose()
@@ -2,20 +2,23 @@
Test rooms API endpoints in the Meet core app: participants management.
"""
# pylint: disable=redefined-outer-name,unused-argument,protected-access
# pylint: disable=redefined-outer-name,unused-argument,protected-access,no-name-in-module,too-many-lines
import random
from unittest import mock
from uuid import uuid4
from django.contrib.auth.models import AnonymousUser
from django.core.exceptions import SuspiciousOperation
from django.urls import reverse
import pytest
from livekit.api import TwirpError
from livekit.api import TwirpError, UpdateParticipantRequest
from livekit.protocol.models import ParticipantInfo
from rest_framework import status
from rest_framework.test import APIClient
from core import utils
from core.factories import RoomFactory, UserFactory, UserResourceAccessFactory
from core.services.lobby import LobbyService
@@ -31,8 +34,8 @@ def mock_livekit_client():
yield mock_client
def test_mute_participant_success(mock_livekit_client):
"""Test successful participant muting."""
def test_mute_participant_success_as_admin(mock_livekit_client):
"""Admins and owners should be able to mute without a LiveKit token."""
client = APIClient()
room = RoomFactory()
user = UserFactory()
@@ -41,10 +44,12 @@ def test_mute_participant_success(mock_livekit_client):
)
client.force_authenticate(user=user)
payload = {"participant_identity": str(uuid4()), "track_sid": "test-track-sid"}
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(url, payload, format="json")
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
)
assert response.status_code == status.HTTP_200_OK
assert response.data == {"status": "success"}
@@ -53,23 +58,131 @@ def test_mute_participant_success(mock_livekit_client):
mock_livekit_client.aclose.assert_called_once()
def test_mute_participant_forbidden_without_access():
"""Test mute participant returns 403 when user lacks room privileges."""
def test_mute_participant_anonymous_no_token_forbidden(mock_livekit_client):
"""Should forbid muting when user is anonymous and no LiveKit token."""
client = APIClient()
room = RoomFactory()
user = UserFactory() # User without UserResourceAccess
client.force_authenticate(user=user)
payload = {"participant_identity": str(uuid4()), "track_sid": "test-track-sid"}
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(url, payload, format="json")
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
)
assert response.status_code == status.HTTP_403_FORBIDDEN
mock_livekit_client.room.mute_published_track.assert_not_called()
def test_mute_participant_with_livekit_token_for_this_room(mock_livekit_client):
"""Should allow muting when the LiveKit token is scoped to this room."""
client = APIClient()
room = RoomFactory()
user = AnonymousUser()
token = utils.generate_token(str(room.id), user, is_admin_or_owner=False)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_200_OK
assert response.data == {"status": "success"}
mock_livekit_client.room.mute_published_track.assert_called_once()
def test_mute_participant_with_livekit_token_for_another_room_forbidden(
mock_livekit_client,
):
"""Should forbid muting when the LiveKit token is scoped to a different room."""
client = APIClient()
target_room = RoomFactory()
other_room = RoomFactory()
user = AnonymousUser()
token = utils.generate_token(str(other_room.id), user, is_admin_or_owner=False)
url = reverse("rooms-mute-participant", kwargs={"pk": target_room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_403_FORBIDDEN
mock_livekit_client.room.mute_published_track.assert_not_called()
def test_mute_participant_authenticated_no_role_no_token_forbidden(mock_livekit_client):
"""Should forbid muting when user has no room role and no LiveKit token."""
client = APIClient()
room = RoomFactory() # everyone_can_mute defaults to True
user = UserFactory() # no UserResourceAccess for this room
client.force_authenticate(user=user)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
)
assert response.status_code == status.HTTP_403_FORBIDDEN
mock_livekit_client.room.mute_published_track.assert_not_called()
def test_mute_participant_everyone_can_mute_disabled_blocks_non_admin(
mock_livekit_client,
):
"""Should forbid muting when everyone_can_mute is False, even with a LiveKit token."""
client = APIClient()
room = RoomFactory(configuration={"everyone_can_mute": False})
user = AnonymousUser()
token = utils.generate_token(str(room.id), user, is_admin_or_owner=False)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_403_FORBIDDEN
mock_livekit_client.room.mute_published_track.assert_not_called()
def test_mute_participant_everyone_can_mute_disabled_allows_admin(mock_livekit_client):
"""Should allow admins and owners to mute when everyone_can_mute is False."""
client = APIClient()
room = RoomFactory(configuration={"everyone_can_mute": False})
user = UserFactory()
UserResourceAccessFactory(
resource=room, user=user, role=random.choice(["administrator", "owner"])
)
client.force_authenticate(user=user)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
)
assert response.status_code == status.HTTP_200_OK
mock_livekit_client.room.mute_published_track.assert_called_once()
def test_mute_participant_invalid_payload():
"""Test mute participant with invalid payload."""
"""Should reject muting when the payload is invalid."""
client = APIClient()
room = RoomFactory()
user = UserFactory()
@@ -78,16 +191,16 @@ def test_mute_participant_invalid_payload():
)
client.force_authenticate(user=user)
payload = {"participant_identity": "invalid-uuid", "track_sid": ""}
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(url, payload, format="json")
response = client.post(
url, {"participant_identity": "invalid-uuid", "track_sid": ""}, format="json"
)
assert response.status_code == status.HTTP_400_BAD_REQUEST
def test_mute_participant_unexpected_twirp_error(mock_livekit_client):
"""Test mute participant when LiveKit API raises TwirpError."""
"""Should return 500 when the LiveKit API raises a TwirpError."""
client = APIClient()
mock_livekit_client.room.mute_published_track.side_effect = TwirpError(
@@ -101,10 +214,12 @@ def test_mute_participant_unexpected_twirp_error(mock_livekit_client):
)
client.force_authenticate(user=user)
payload = {"participant_identity": str(uuid4()), "track_sid": "test-track-sid"}
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(url, payload, format="json")
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
)
assert response.status_code == status.HTTP_500_INTERNAL_SERVER_ERROR
assert response.data == {"error": "Failed to mute participant"}
@@ -112,6 +227,282 @@ def test_mute_participant_unexpected_twirp_error(mock_livekit_client):
mock_livekit_client.aclose.assert_called_once()
def test_mute_participant_participant_not_found(mock_livekit_client):
"""Should return 404 when the participant does not exist in the room."""
client = APIClient()
mock_livekit_client.room.mute_published_track.side_effect = TwirpError(
msg="participant does not exist", code="not_found", status=404
)
room = RoomFactory()
user = UserFactory()
UserResourceAccessFactory(
resource=room, user=user, role=random.choice(["administrator", "owner"])
)
client.force_authenticate(user=user)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
)
assert response.status_code == status.HTTP_404_NOT_FOUND
assert response.data == {"error": "Participant not found"}
mock_livekit_client.aclose.assert_called_once()
def test_mute_participant_management_exception(mock_livekit_client):
"""Should return 500 when ParticipantsManagement raises an unexpected error."""
client = APIClient()
mock_livekit_client.room.mute_published_track.side_effect = TwirpError(
msg="boom", code="internal", status=503
)
room = RoomFactory()
user = UserFactory()
UserResourceAccessFactory(
resource=room, user=user, role=random.choice(["administrator", "owner"])
)
client.force_authenticate(user=user)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
)
assert response.status_code == status.HTTP_500_INTERNAL_SERVER_ERROR
assert response.data == {"error": "Failed to mute participant"}
mock_livekit_client.aclose.assert_called_once()
def test_mute_participant_admin_with_token_for_this_room(mock_livekit_client):
"""Should allow muting when user is admin and LiveKit token is scoped to this room."""
client = APIClient()
room = RoomFactory()
user = UserFactory()
UserResourceAccessFactory(
resource=room, user=user, role=random.choice(["administrator", "owner"])
)
# Token identity matches the admin user so LiveKitTokenAuthentication
# resolves request.user back to the admin.
token = utils.generate_token(str(room.id), user, is_admin_or_owner=True)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_200_OK
assert response.data == {"status": "success"}
mock_livekit_client.room.mute_published_track.assert_called_once()
def test_mute_participant_admin_with_token_for_another_room(mock_livekit_client):
"""Should not allow muting when user is admin and the LiveKit token is for another room."""
client = APIClient()
target_room = RoomFactory()
other_room = RoomFactory()
user = UserFactory()
UserResourceAccessFactory(
resource=target_room,
user=user,
role=random.choice(["administrator", "owner"]),
)
# Token is scoped to a DIFFERENT room, and admin status must only be
# honored when established via session, never via a LiveKit
# token, which can be replayed off-host.
token = utils.generate_token(str(other_room.id), user, is_admin_or_owner=True)
url = reverse("rooms-mute-participant", kwargs={"pk": target_room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_403_FORBIDDEN
assert response.data == {
"detail": "You do not have permission to perform this action."
}
mock_livekit_client.room.mute_published_track.assert_not_called()
def test_mute_participant_admin_token_replayed_does_not_grant_admin(
mock_livekit_client,
):
"""Should forbid muting when a LiveKit token issued for an admin is passed without a session."""
client = APIClient()
room = RoomFactory(configuration={"everyone_can_mute": False})
admin_user = UserFactory()
UserResourceAccessFactory(
resource=room,
user=admin_user,
role=random.choice(["administrator", "owner"]),
)
# The token is the only credential.
token = utils.generate_token(str(room.id), admin_user, is_admin_or_owner=True)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_403_FORBIDDEN
mock_livekit_client.room.mute_published_track.assert_not_called()
def test_mute_participant_livekit_token_triggers_presence_check(mock_livekit_client):
"""Should check participant presence when authenticated via LiveKit token only."""
client = APIClient()
room = RoomFactory()
user = AnonymousUser()
token = utils.generate_token(str(room.id), user, is_admin_or_owner=False)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_200_OK
# Presence is verified against LiveKit before the mute is issued.
mock_livekit_client.room.get_participant.assert_called_once()
mock_livekit_client.room.mute_published_track.assert_called_once()
def test_mute_participant_livekit_token_presence_check_returns_participant(
mock_livekit_client,
):
"""Should mute when the authentified participant is currently in the room."""
client = APIClient()
room = RoomFactory()
# Simulate LiveKit confirming the caller is currently in the room.
# State != DISCONNECTED (3) means present.
mock_livekit_client.room.get_participant.return_value = ParticipantInfo(
identity="caller-identity",
state=ParticipantInfo.State.ACTIVE,
)
user = AnonymousUser()
token = utils.generate_token(str(room.id), user, is_admin_or_owner=False)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_200_OK
assert response.data == {"status": "success"}
mock_livekit_client.room.get_participant.assert_called_once()
mock_livekit_client.room.mute_published_track.assert_called_once()
def test_mute_participant_livekit_token_presence_check_participant_not_found(
mock_livekit_client,
):
"""Should not mute when the authentified participant is not found."""
client = APIClient()
room = RoomFactory()
mock_livekit_client.room.get_participant.side_effect = TwirpError(
msg="participant does not exist", code="not_found", status=404
)
user = AnonymousUser()
token = utils.generate_token(str(room.id), user, is_admin_or_owner=False)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_403_FORBIDDEN
assert response.data == {"error": "Could not verify caller presence"}
mock_livekit_client.room.get_participant.assert_called_once()
# The presence check failed, so we never reach the mute call.
mock_livekit_client.room.mute_published_track.assert_not_called()
def test_mute_participant_livekit_token_presence_check_twirp_error_forbidden(
mock_livekit_client,
):
"""Should not mute when the presence check fail."""
client = APIClient()
room = RoomFactory()
mock_livekit_client.room.get_participant.side_effect = TwirpError(
msg="an error occured", code="not_found", status=500
)
user = AnonymousUser()
token = utils.generate_token(str(room.id), user, is_admin_or_owner=False)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
HTTP_AUTHORIZATION=f"Bearer {token}",
)
assert response.status_code == status.HTTP_403_FORBIDDEN
assert response.data == {"error": "Could not verify caller presence"}
mock_livekit_client.room.get_participant.assert_called_once()
# The presence check failed, so we never reach the mute call.
mock_livekit_client.room.mute_published_track.assert_not_called()
def test_mute_participant_session_auth_skips_presence_check(mock_livekit_client):
"""Should not check presence of the participant when authentified with a session cookie."""
client = APIClient()
room = RoomFactory()
user = UserFactory()
UserResourceAccessFactory(
resource=room, user=user, role=random.choice(["administrator", "owner"])
)
client.force_authenticate(user=user)
url = reverse("rooms-mute-participant", kwargs={"pk": room.id})
response = client.post(
url,
{"participant_identity": str(uuid4()), "track_sid": "test-track-sid"},
format="json",
)
assert response.status_code == status.HTTP_200_OK
# Session auth has no LiveKit identity to verify against, so the
# stop-gap presence check is skipped.
mock_livekit_client.room.get_participant.assert_not_called()
mock_livekit_client.room.mute_published_track.assert_called_once()
def test_update_participant_success(mock_livekit_client):
"""Test successful participant update."""
client = APIClient()
@@ -130,8 +521,8 @@ def test_update_participant_success(mock_livekit_client):
"can_publish": True,
"can_publish_data": True,
"can_publish_sources": [
"CAMERA",
"MICROPHONE",
"camera",
"microphone",
],
"can_update_metadata": True,
"can_subscribe_metrics": True,
@@ -158,8 +549,8 @@ def test_update_participant_success(mock_livekit_client):
{"can_publish_data": True},
{
"can_publish_sources": [
"CAMERA",
"MICROPHONE",
"camera",
"microphone",
]
},
{"can_update_metadata": True},
@@ -190,9 +581,41 @@ def test_update_participant_permission_fields_are_optional(
assert response.data == {"status": "success"}
mock_livekit_client.room.update_participant.assert_called_once()
(request_arg,), _ = mock_livekit_client.room.update_participant.call_args
assert isinstance(request_arg, UpdateParticipantRequest)
mock_livekit_client.aclose.assert_called_once()
def test_update_participant_permission_fields_invalid_case(mock_livekit_client):
"""Should raise bad request when can_publish_sources is uppercase."""
client = APIClient()
room = RoomFactory()
user = UserFactory()
UserResourceAccessFactory(
resource=room, user=user, role=random.choice(["administrator", "owner"])
)
client.force_authenticate(user=user)
payload = {
"participant_identity": str(uuid4()),
"permission": {
"can_publish_sources": [
"CAMERA",
"microphone",
]
},
}
url = reverse("rooms-update-participant", kwargs={"pk": room.id})
response = client.post(url, payload, format="json")
assert response.status_code == status.HTTP_400_BAD_REQUEST
mock_livekit_client.room.update_participant.assert_not_called()
mock_livekit_client.aclose.assert_not_called()
@pytest.mark.parametrize(
"value,permission_key",
[
@@ -28,6 +28,7 @@ def test_api_rooms_retrieve_anonymous_private_pk():
assert response.status_code == 200
assert response.json() == {
"configuration": {},
"access_level": "restricted",
"id": str(room.id),
"is_administrable": False,
@@ -47,6 +48,7 @@ def test_api_rooms_retrieve_anonymous_trusted_pk():
assert response.status_code == 200
assert response.json() == {
"configuration": {},
"access_level": "trusted",
"id": str(room.id),
"is_administrable": False,
@@ -65,6 +67,7 @@ def test_api_rooms_retrieve_anonymous_private_pk_no_dashes():
assert response.status_code == 200
assert response.json() == {
"configuration": {},
"access_level": "restricted",
"id": str(room.id),
"is_administrable": False,
@@ -81,6 +84,7 @@ def test_api_rooms_retrieve_anonymous_private_slug():
assert response.status_code == 200
assert response.json() == {
"configuration": {},
"access_level": "restricted",
"id": str(room.id),
"is_administrable": False,
@@ -97,6 +101,7 @@ def test_api_rooms_retrieve_anonymous_private_slug_not_normalized():
assert response.status_code == 200
assert response.json() == {
"configuration": {},
"access_level": "restricted",
"id": str(room.id),
"is_administrable": False,
@@ -200,6 +205,7 @@ def test_api_rooms_retrieve_anonymous_public(mock_token):
assert response.status_code == 200
expected_name = f"{room.id!s}"
assert response.json() == {
"configuration": {},
"access_level": str(room.access_level),
"id": str(room.id),
"is_administrable": False,
@@ -246,6 +252,7 @@ def test_api_rooms_retrieve_authenticated_public(mock_token):
expected_name = f"{room.id!s}"
assert response.json() == {
"configuration": {"can_publish_sources": ["camera"]},
"access_level": str(room.access_level),
"id": str(room.id),
"is_administrable": False,
@@ -297,6 +304,7 @@ def test_api_rooms_retrieve_authenticated_trusted(mock_token):
expected_name = f"{room.id!s}"
assert response.json() == {
"configuration": {},
"access_level": str(room.access_level),
"id": str(room.id),
"is_administrable": False,
@@ -338,6 +346,7 @@ def test_api_rooms_retrieve_authenticated():
assert response.status_code == 200
assert response.json() == {
"configuration": {},
"access_level": "restricted",
"id": str(room.id),
"is_administrable": False,
@@ -383,6 +392,7 @@ def test_api_rooms_retrieve_members(mock_token, django_assert_num_queries, setti
expected_name = str(room.id)
assert content_dict == {
"configuration": {"can_publish_sources": ["camera"]},
"access_level": str(room.access_level),
"id": str(room.id),
"is_administrable": False,
@@ -3,12 +3,18 @@ Test rooms API endpoints in the Meet core app: update.
"""
import random
from unittest.mock import patch
import pytest
from rest_framework.test import APIClient
from ...factories import RoomFactory, UserFactory
from ...models import RoomAccessLevel
from ...services.room_management import (
RoomManagement,
RoomManagementException,
RoomNotFoundException,
)
pytestmark = pytest.mark.django_db
@@ -79,12 +85,14 @@ def test_api_rooms_update_members():
assert room.configuration == {}
def test_api_rooms_update_administrators():
"""Administrators or owners of a room should be allowed to update it."""
@patch.object(RoomManagement, "update_metadata")
def test_api_rooms_update_administrators(mock_update_metadata):
"""Should sync LiveKit metadata when both configuration and access level change."""
user = UserFactory()
room = RoomFactory(
access_level=RoomAccessLevel.RESTRICTED,
users=[(user, random.choice(["administrator", "owner"]))],
configuration={"can_publish_sources": ["camera"]},
)
client = APIClient()
client.force_login(user)
@@ -106,11 +114,120 @@ def test_api_rooms_update_administrators():
assert room.access_level == RoomAccessLevel.PUBLIC
assert room.configuration == {"can_publish_sources": ["camera", "microphone"]}
mock_update_metadata.assert_called_once_with(
room_name=str(room.id),
metadata={
"access_level": "public",
"configuration": {"can_publish_sources": ["camera", "microphone"]},
},
)
@patch.object(RoomManagement, "update_metadata")
def test_api_rooms_update_administrators_configuration_only(mock_update_metadata):
"""Should sync LiveKit metadata when only configuration changes."""
user = UserFactory()
room = RoomFactory(
access_level=RoomAccessLevel.RESTRICTED,
users=[(user, random.choice(["administrator", "owner"]))],
configuration={},
)
client = APIClient()
client.force_login(user)
response = client.put(
f"/api/v1.0/rooms/{room.id!s}/",
{
"name": "New name",
"slug": "should-be-ignored",
"configuration": {"can_publish_sources": ["camera", "microphone"]},
},
format="json",
)
assert response.status_code == 200
room.refresh_from_db()
assert room.name == "New name"
assert room.slug == "new-name"
assert room.access_level == RoomAccessLevel.RESTRICTED
assert room.configuration == {"can_publish_sources": ["camera", "microphone"]}
mock_update_metadata.assert_called_once_with(
room_name=str(room.id),
metadata={
"access_level": "restricted",
"configuration": {"can_publish_sources": ["camera", "microphone"]},
},
)
@patch.object(RoomManagement, "update_metadata")
def test_api_rooms_update_administrators_access_level_only(mock_update_metadata):
"""Should sync LiveKit metadata when only access level changes."""
user = UserFactory()
room = RoomFactory(
access_level=RoomAccessLevel.RESTRICTED,
users=[(user, random.choice(["administrator", "owner"]))],
configuration={"can_publish_sources": ["camera"]},
)
client = APIClient()
client.force_login(user)
response = client.put(
f"/api/v1.0/rooms/{room.id!s}/",
{
"name": "New name",
"access_level": RoomAccessLevel.PUBLIC,
},
format="json",
)
assert response.status_code == 200
room.refresh_from_db()
assert room.name == "New name"
assert room.slug == "new-name"
assert room.access_level == RoomAccessLevel.PUBLIC
assert room.configuration == {"can_publish_sources": ["camera"]}
mock_update_metadata.assert_called_once_with(
room_name=str(room.id),
metadata={
"access_level": "public",
"configuration": {"can_publish_sources": ["camera"]},
},
)
@patch.object(RoomManagement, "update_metadata")
def test_api_rooms_update_administrators_name_only(mock_update_metadata):
"""Should not sync LiveKit metadata when neither configuration nor access level changes."""
user = UserFactory()
room = RoomFactory(
name="Old name",
access_level=RoomAccessLevel.PUBLIC,
configuration={"can_publish_sources": ["camera"]},
users=[(user, random.choice(["administrator", "owner"]))],
)
client = APIClient()
client.force_login(user)
response = client.patch(
f"/api/v1.0/rooms/{room.id!s}/",
{"name": "New name"},
format="json",
)
assert response.status_code == 200
room.refresh_from_db()
assert room.name == "New name"
assert room.slug == "new-name"
# Unrelated fields untouched
assert room.access_level == RoomAccessLevel.PUBLIC
assert room.configuration == {"can_publish_sources": ["camera"]}
mock_update_metadata.assert_not_called()
@pytest.mark.parametrize(
"configuration",
[
{},
{"can_publish_sources": ["camera", "microphone"]},
{
"can_publish_sources": [
@@ -122,12 +239,17 @@ def test_api_rooms_update_administrators():
},
{"can_publish_sources": []},
{"can_publish_sources": None},
{"can_publish_sources": None, "everyone_can_mute": True},
{"can_publish_sources": None, "everyone_can_mute": False},
{"can_publish_sources": None, "everyone_can_mute": "yes"},
{"can_publish_sources": None, "everyone_can_mute": "1"},
],
)
def test_api_rooms_update_configuration_valid(configuration):
@patch.object(RoomManagement, "update_metadata")
def test_api_rooms_update_configuration_valid(mock_update_metadata, configuration):
"""Administrators should be allowed to set valid configurations."""
user = UserFactory()
room = RoomFactory(users=[(user, "owner")])
room = RoomFactory(users=[(user, "owner")], configuration={})
client = APIClient()
client.force_login(user)
@@ -140,6 +262,28 @@ def test_api_rooms_update_configuration_valid(configuration):
room.refresh_from_db()
assert room.configuration == configuration
mock_update_metadata.assert_called_once()
@patch.object(RoomManagement, "update_metadata")
def test_api_rooms_update_configuration_unchanged_empty(mock_update_metadata):
"""Should not sync LiveKit metadata when patching an already empty configuration."""
user = UserFactory()
room = RoomFactory(users=[(user, "owner")], configuration={})
client = APIClient()
client.force_login(user)
response = client.patch(
f"/api/v1.0/rooms/{room.id!s}/",
{"configuration": {}},
format="json",
)
assert response.status_code == 200
room.refresh_from_db()
assert room.configuration == {}
mock_update_metadata.assert_not_called()
def test_api_rooms_update_configuration_extra_keys_rejected():
"""Extra keys in configuration should be rejected."""
@@ -198,6 +342,24 @@ def test_api_rooms_update_configuration_wrong_type():
assert room.configuration == {}
@pytest.mark.parametrize("invalid_value", ["test", [], {}])
def test_api_rooms_update_configuration_everyone_can_mute_wrong_type(invalid_value):
"""everyone_can_mute values with wrong types should be rejected."""
user = UserFactory()
room = RoomFactory(users=[(user, "owner")])
client = APIClient()
client.force_login(user)
response = client.patch(
f"/api/v1.0/rooms/{room.id!s}/",
{"configuration": {"everyone_can_mute": invalid_value}},
format="json",
)
assert response.status_code == 400
room.refresh_from_db()
assert room.configuration == {}
def test_api_rooms_update_administrators_of_another():
"""
Being administrator or owner of a room should not grant authorization to update
@@ -217,3 +379,61 @@ def test_api_rooms_update_administrators_of_another():
other_room.refresh_from_db()
assert other_room.name == "Old name"
assert other_room.slug == "old-name"
@patch.object(RoomManagement, "update_metadata", side_effect=RoomNotFoundException)
def test_api_rooms_update_livekit_room_not_found(mock_update_metadata):
"""Should not fail the API request when the LiveKit room does not exist yet."""
user = UserFactory()
room = RoomFactory(
users=[(user, random.choice(["administrator", "owner"]))],
configuration={},
)
client = APIClient()
client.force_login(user)
response = client.patch(
f"/api/v1.0/rooms/{room.id!s}/",
{"configuration": {"can_publish_sources": ["camera"]}},
format="json",
)
assert response.status_code == 200
room.refresh_from_db()
assert room.configuration == {"can_publish_sources": ["camera"]}
mock_update_metadata.assert_called_once_with(
room_name=str(room.id),
metadata={
"access_level": room.access_level,
"configuration": {"can_publish_sources": ["camera"]},
},
)
@patch.object(RoomManagement, "update_metadata", side_effect=RoomManagementException)
def test_api_rooms_update_livekit_sync_failure(mock_update_metadata):
"""Should not fail the API request when the LiveKit metadata sync fails."""
user = UserFactory()
room = RoomFactory(
users=[(user, random.choice(["administrator", "owner"]))],
configuration={},
)
client = APIClient()
client.force_login(user)
response = client.patch(
f"/api/v1.0/rooms/{room.id!s}/",
{"configuration": {"can_publish_sources": ["camera"]}},
format="json",
)
assert response.status_code == 200
room.refresh_from_db()
assert room.configuration == {"can_publish_sources": ["camera"]}
mock_update_metadata.assert_called_once_with(
room_name=str(room.id),
metadata={
"access_level": room.access_level,
"configuration": {"can_publish_sources": ["camera"]},
},
)
+5 -1
View File
@@ -121,7 +121,11 @@ def generate_token(
.with_identity(identity)
.with_name(username or default_username)
.with_attributes(
{"color": color, "room_admin": "true" if is_admin_or_owner else "false"}
{
"color": color,
"room_admin": "true" if is_admin_or_owner else "false",
"is_authenticated": not user.is_anonymous,
}
)
)
+1 -1
View File
@@ -7,7 +7,7 @@ build-backend = "uv_build"
[project]
name = "meet"
version = "1.15.0"
version = "1.16.0"
authors = [{ "name" = "DINUM", "email" = "dev@mail.numerique.gouv.fr" }]
classifiers = [
"Development Status :: 5 - Production/Stable",
+1 -1
View File
@@ -1173,7 +1173,7 @@ wheels = [
[[package]]
name = "meet"
version = "1.15.0"
version = "1.16.0"
source = { editable = "." }
dependencies = [
{ name = "aiohttp" },
+421 -462
View File
File diff suppressed because it is too large Load Diff
+3 -3
View File
@@ -1,7 +1,7 @@
{
"name": "meet",
"private": true,
"version": "1.15.0",
"version": "1.16.0",
"type": "module",
"scripts": {
"dev": "panda codegen && vite",
@@ -47,7 +47,7 @@
"wouter": "3.9.0"
},
"devDependencies": {
"@pandacss/dev": "1.8.2",
"@pandacss/dev": "1.11.1",
"@tanstack/eslint-plugin-query": "5.91.4",
"@tanstack/react-query-devtools": "5.91.3",
"@types/humanize-duration": "3.27.4",
@@ -62,7 +62,7 @@
"eslint-plugin-jsx-a11y": "6.10.2",
"eslint-plugin-react-hooks": "5.2.0",
"eslint-plugin-react-refresh": "0.4.20",
"postcss": "8.5.10",
"postcss": "8.5.14",
"prettier": "3.8.1",
"typescript": "5.8.3",
"vite": "7.3.2",
+3 -1
View File
@@ -2,6 +2,8 @@ import { fetchApi } from './fetchApi'
import { keys } from './queryKeys'
import { useQuery } from '@tanstack/react-query'
import { RecordingMode } from '@/features/recording'
import { Track } from 'livekit-client'
import Source = Track.Source
export interface ApiConfig {
analytics?: {
@@ -50,7 +52,7 @@ export interface ApiConfig {
url: string
force_wss_protocol: boolean
enable_firefox_proxy_workaround: boolean
default_sources: string[]
default_sources: Source[]
}
transcription_destination?: string
}
@@ -12,7 +12,7 @@ const controlBarRegion = cva({
variants: {
mobile: {
true: {
justifyContent: 'space-between',
justifyContent: 'center',
width: '330px',
},
},
@@ -1,3 +1,6 @@
import { Track } from 'livekit-client'
import Source = Track.Source
export type ApiLiveKit = {
url: string
room: string
@@ -10,6 +13,11 @@ export enum ApiAccessLevel {
RESTRICTED = 'restricted',
}
export type RoomConfiguration = {
can_publish_sources?: Source[] | null
everyone_can_mute?: boolean | null
}
export type ApiRoom = {
id: string
name: string
@@ -18,7 +26,5 @@ export type ApiRoom = {
is_administrable: boolean
access_level: ApiAccessLevel
livekit?: ApiLiveKit
configuration?: {
[key: string]: string | number | boolean | string[]
}
configuration?: RoomConfiguration
}
@@ -6,44 +6,74 @@ import {
NotificationType,
} from '@/features/notifications'
import { fetchApi } from '@/api/fetchApi'
import { useIsAdminOrOwner } from '../livekit/hooks/useIsAdminOrOwner'
import { useCallback } from 'react'
export const useMuteParticipant = () => {
const data = useRoomData()
const apiRoomData = useRoomData()
const { notifyParticipants } = useNotifyParticipants()
const isAdminOrOwner = useIsAdminOrOwner()
const muteParticipant = async (participant: Participant) => {
if (!data?.id) {
throw new Error('Room id is not available')
}
const trackSid = participant.getTrackPublication(
Source.Microphone
)?.trackSid
const muteParticipant = useCallback(
async (participant: Participant) => {
if (!apiRoomData?.livekit?.room) {
throw new Error('Room id is not available')
}
if (!trackSid) {
return
}
const trackSid = participant.getTrackPublication(
Source.Microphone
)?.trackSid
try {
const response = await fetchApi(`rooms/${data.id}/mute-participant/`, {
method: 'POST',
body: JSON.stringify({
participant_identity: participant.identity,
track_sid: trackSid,
}),
})
if (!trackSid) {
return
}
await notifyParticipants({
type: NotificationType.ParticipantMuted,
destinationIdentities: [participant.identity],
})
// Guard against undefined token for non-admin users
if (!isAdminOrOwner && !apiRoomData.livekit.token) {
console.error('Cannot mute participant: missing auth token')
return
}
const headers = !isAdminOrOwner
? { Authorization: `Bearer ${apiRoomData.livekit.token}` }
: undefined
let response
try {
response = await fetchApi(
`rooms/${apiRoomData.livekit.room}/mute-participant/`,
{
method: 'POST',
headers,
body: JSON.stringify({
participant_identity: participant.identity,
track_sid: trackSid,
}),
}
)
} catch (error) {
console.error(
`Failed to mute participant ${participant.identity}: ${error instanceof Error ? error.message : 'Unknown error'}`
)
return
}
try {
await notifyParticipants({
type: NotificationType.ParticipantMuted,
destinationIdentities: [participant.identity],
})
} catch (e) {
console.error(
`Failed to notify muted participant ${participant.identity}: ${e}`
)
}
return response
} catch (error) {
console.error(
`Failed to mute participant ${participant.identity}: ${error instanceof Error ? error.message : 'Unknown error'}`
)
}
}
},
[apiRoomData, isAdminOrOwner, notifyParticipants]
)
return { muteParticipant }
}
@@ -8,7 +8,7 @@ export const useParticipantPermissions = () => {
const updateParticipantPermissions = async (
participant: Participant,
sources: Array<Source>
sources: Source[]
) => {
if (!data?.id) {
throw new Error('Room id is not available')
@@ -20,7 +20,7 @@ export const useParticipantPermissions = () => {
can_update_metadata: participant.permissions?.canUpdateMetadata,
can_subscribe_metrics: participant.permissions?.canSubscribeMetrics,
can_publish: sources.length > 0,
can_publish_sources: sources.map((source) => source.toUpperCase()),
can_publish_sources: sources,
}
try {
@@ -9,7 +9,8 @@ import { queryClient } from '@/api/queryClient'
import { keys } from '@/api/queryKeys'
import { useQuery } from '@tanstack/react-query'
import { useParams } from 'wouter'
import { usePublishSourcesManager } from '@/features/rooms/livekit/hooks/usePublishSourcesManager'
import { usePublishSourcesManager } from '../hooks/usePublishSourcesManager'
import { usePermissionsManager } from '../hooks/usePermissionsManager'
export const Admin = () => {
const { t } = useTranslation('rooms', { keyPrefix: 'admin' })
@@ -38,6 +39,8 @@ export const Admin = () => {
isScreenShareEnabled,
} = usePublishSourcesManager()
const { toggleMuting, isMutingEnabled } = usePermissionsManager()
return (
<Div
display="flex"
@@ -130,6 +133,17 @@ export const Admin = () => {
fullWidth: true,
}}
/>
<Field
type="switch"
label={t('moderation.mute.label')}
description={t('moderation.mute.description')}
isSelected={isMutingEnabled}
onChange={toggleMuting}
wrapperProps={{
noMargin: true,
fullWidth: true,
}}
/>
</div>
</div>
<div
@@ -1,7 +1,13 @@
import { useIsAdminOrOwner } from './useIsAdminOrOwner'
import { Participant } from 'livekit-client'
import { useRoomData } from '@/features/rooms/livekit/hooks/useRoomData'
export const useCanMute = (participant: Participant) => {
const apiRoomData = useRoomData()
const isAdminOrOwner = useIsAdminOrOwner()
return participant.isLocal || isAdminOrOwner
return (
participant.isLocal ||
isAdminOrOwner ||
apiRoomData?.configuration?.everyone_can_mute !== false
)
}
@@ -0,0 +1,46 @@
import { usePatchRoom } from '@/features/rooms/api/patchRoom'
import { useRoomData } from '@/features/rooms/livekit/hooks/useRoomData'
import { useCallback } from 'react'
import { queryClient } from '@/api/queryClient'
import { keys } from '@/api/queryKeys'
export const usePermissionsManager = () => {
const { mutateAsync: patchRoom } = usePatchRoom()
const data = useRoomData()
const configuration = data?.configuration
const roomId = data?.slug
const isMutingEnabled = configuration?.everyone_can_mute ?? true
const toggleMuting = useCallback(
async (enabled: boolean) => {
if (!roomId) return
try {
const newConfiguration = {
...configuration,
everyone_can_mute: enabled,
}
const room = await patchRoom({
roomId,
room: { configuration: newConfiguration },
})
queryClient.setQueryData([keys.room, roomId], room)
return { configuration: newConfiguration }
} catch (error) {
console.error('Failed to update muting permission:', error)
return { success: false, error }
}
},
[configuration, roomId, patchRoom]
)
return {
toggleMuting,
isMutingEnabled,
}
}
@@ -39,10 +39,6 @@ export const usePublishSourcesManager = () => {
const { notifyParticipants } = useNotifyParticipants()
const defaultSources = configData?.livekit?.default_sources?.map((source) => {
return source as Source
})
// The name can be misleading—use the slug instead to ensure the correct React Query key is updated.
const roomId = data?.slug
@@ -54,16 +50,16 @@ export const usePublishSourcesManager = () => {
)
const currentSources = useMemo(() => {
const defaultSources = configData?.livekit?.default_sources ?? []
if (
configuration?.can_publish_sources == undefined ||
!Array.isArray(configuration?.can_publish_sources)
) {
return defaultSources
}
return configuration.can_publish_sources.map((source) => {
return source as Source
})
}, [defaultSources, configuration?.can_publish_sources])
return configuration.can_publish_sources
}, [configData, configuration?.can_publish_sources])
const updateSource = useCallback(
async (sources: Source[], enabled: boolean) => {
@@ -78,7 +74,7 @@ export const usePublishSourcesManager = () => {
const newConfiguration = {
...configuration,
can_publish_sources: newSources as string[],
can_publish_sources: newSources,
}
const room = await patchRoom({
@@ -0,0 +1,86 @@
// features/rooms/hooks/useSyncLiveKitMetadata.ts
import { useEffect } from 'react'
import { RoomEvent } from 'livekit-client'
import { queryClient } from '@/api/queryClient'
import { keys } from '@/api/queryKeys'
import {
ApiAccessLevel,
ApiRoom,
RoomConfiguration,
} from '@/features/rooms/api/ApiRoom'
import { useRoomContext } from '@livekit/components-react'
import { useRoomData } from './useRoomData'
/**
* Shape of the LiveKit room metadata blob pushed by the backend.
* Matches RoomManagement.update_metadata → {"configuration": room.configuration}
*/
type RoomLiveKitMetadata = {
configuration?: RoomConfiguration
access_level?: ApiAccessLevel
}
const parseMetadata = (raw: string | undefined): RoomLiveKitMetadata | null => {
if (!raw) return null
try {
return JSON.parse(raw) as RoomLiveKitMetadata
} catch {
console.warn('useSyncLiveKitMetadata: failed to parse room metadata')
return null
}
}
/**
* Sync LiveKit room metadata into the React Query cache.
*
* The backend pushes room configuration into LiveKit's room metadata
* whenever it changes. This hook listens for those changes and patches
* the ApiRoom cache so every `useRoomData()`
* consumer sees the fresh value automatically.
*
* Mount once, at the level where the LiveKit Room instance lives.
*/
export const useSyncLiveKitMetadata = () => {
const room = useRoomContext()
const roomData = useRoomData()
const roomSlug = roomData?.slug
useEffect(() => {
if (!room || !roomSlug) return
const applyMetadata = (raw: string | undefined) => {
const parsed = parseMetadata(raw)
if (!parsed) return
queryClient.setQueryData<ApiRoom>([keys.room, roomSlug], (prev) => {
if (!prev) return prev
const nextConfiguration = parsed.configuration ?? prev.configuration
const nextAccessLevel = parsed.access_level ?? prev.access_level
if (
nextConfiguration === prev.configuration &&
nextAccessLevel === prev.access_level
) {
return prev
}
return {
...prev,
configuration: nextConfiguration,
access_level: nextAccessLevel,
}
})
}
// Apply whatever metadata is currently set (covers the case where we
// joined the room AFTER the last metadata change, so no event will fire).
applyMetadata(room.metadata)
const handler = (raw: string) => applyMetadata(raw)
room.on(RoomEvent.RoomMetadataChanged, handler)
return () => {
room.off(RoomEvent.RoomMetadataChanged, handler)
}
}, [room, roomSlug])
}
@@ -32,6 +32,7 @@ import { useRegisterKeyboardShortcut } from '@/features/shortcuts/useRegisterKey
import { useSettingsDialog } from '@/features/settings'
import { SettingsDialogExtendedKey } from '@/features/settings/type'
import { useVideoResolutionSubscription } from '../hooks/useVideoResolutionSubscription'
import { useSyncLiveKitMetadata } from '../hooks/useSyncLiveKitMetadata'
import { SettingsDialogProvider } from '@/features/settings/components/SettingsDialogProvider'
import { IsIdleDisconnectModal } from '../components/IsIdleDisconnectModal'
import { getParticipantName } from '@/features/rooms/utils/getParticipantName'
@@ -90,6 +91,7 @@ export function VideoConference({ ...props }: VideoConferenceProps) {
useConnectionObserver()
useRoomPageTitle()
useVideoResolutionSubscription()
useSyncLiveKitMetadata()
useRegisterKeyboardShortcut({
id: 'open-shortcuts',
@@ -75,15 +75,15 @@ const useTranscriptionState = () => {
const segment = segments[0]
setTranscriptionSegments((prevSegments) => {
const existingIndex = prevSegments.findIndex(
(s: TranscriptionSegmentWithParticipant) => s.id === segment.id
)
if (existingIndex === -1) {
return [...prevSegments, { participant, ...segment }]
}
const next = prevSegments.slice()
next[existingIndex] = { ...next[existingIndex], ...segment }
return next
const existingSegmentIds = new Set(prevSegments.map((s) => s.id))
if (existingSegmentIds.has(segment.id)) return prevSegments
return [
...prevSegments,
{
participant: participant,
...segment,
},
]
})
}
+4
View File
@@ -528,6 +528,10 @@
"screenshare": {
"label": "Bildschirm teilen",
"description": "Wenn du diese Option deaktivierst, können Teilnehmende ihren Bildschirm nicht mehr teilen. Laufende Bildschirmfreigaben werden sofort beendet."
},
"mute": {
"label": "Andere stummschalten",
"description": "Wenn deaktiviert, können Teilnehmer andere Teilnehmer nicht mehr stummschalten."
}
}
},
+4
View File
@@ -527,6 +527,10 @@
"screenshare": {
"label": "Share their screen",
"description": "Disabling this option will prevent participants from sharing their screen, and any ongoing screen sharing will be stopped immediately."
},
"mute": {
"label": "Mute others",
"description": "When disabled, participants will no longer be able to mute other participants."
}
}
},
+4
View File
@@ -527,6 +527,10 @@
"screenshare": {
"label": "Partager leur écran",
"description": "En désactivant cette option, les participants ne pourront plus partager leur écran et tout partage en cours sera immédiatement interrompu."
},
"mute": {
"label": "Muter les autres",
"description": "En désactivant cette option, les participants ne pourront plus muter d'autres participants."
}
}
},
+4
View File
@@ -527,6 +527,10 @@
"screenshare": {
"label": "Hun scherm delen",
"description": "Als u deze optie uitschakelt, kunnen deelnemers hun scherm niet meer delen en wordt elke lopende schermdeling onmiddellijk gestopt."
},
"mute": {
"label": "Anderen dempen",
"description": "Wanneer uitgeschakeld, kunnen deelnemers andere deelnemers niet meer dempen."
}
}
},
+1 -1
View File
@@ -59,7 +59,7 @@ export const CAPTION_FONT_COLOR_VALUES: Record<CaptionColor, string> = {
}
export const CAPTION_BACKGROUND_COLOR_VALUES: Record<CaptionColor, string> = {
default: 'rgba(0, 0, 0, 0.75)',
default: 'transparent',
black: 'rgba(0, 0, 0, 0.75)',
white: 'rgba(255, 255, 255, 0.75)',
blue: 'rgba(0, 0, 255, 0.75)',
+5 -5
View File
@@ -1,12 +1,12 @@
{
"name": "mail_mjml",
"version": "1.15.0",
"version": "1.16.0",
"lockfileVersion": 3,
"requires": true,
"packages": {
"": {
"name": "mail_mjml",
"version": "1.15.0",
"version": "1.16.0",
"license": "MIT",
"dependencies": {
"@html-to/text-cli": "0.5.4",
@@ -1588,9 +1588,9 @@
}
},
"node_modules/semver": {
"version": "7.7.4",
"resolved": "https://registry.npmjs.org/semver/-/semver-7.7.4.tgz",
"integrity": "sha512-vFKC2IEtQnVhpT78h1Yp8wzwrf8CM+MzKMHGJZfBtzhZNycRFnXsHk6E5TxIkkMsgNS7mdX3AGB7x2QM2di4lA==",
"version": "7.8.0",
"resolved": "https://registry.npmjs.org/semver/-/semver-7.8.0.tgz",
"integrity": "sha512-AcM7dV/5ul4EekoQ29Agm5vri8JNqRyj39o0qpX6vDF2GZrtutZl5RwgD1XnZjiTAfncsJhMI48QQH3sN87YNA==",
"license": "ISC",
"bin": {
"semver": "bin/semver.js"
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "mail_mjml",
"version": "1.15.0",
"version": "1.16.0",
"description": "An util to generate html and text django's templates from mjml templates",
"type": "module",
"dependencies": {
+2 -2
View File
@@ -1,12 +1,12 @@
{
"name": "sdk",
"version": "1.15.0",
"version": "1.16.0",
"lockfileVersion": 3,
"requires": true,
"packages": {
"": {
"name": "sdk",
"version": "1.15.0",
"version": "1.16.0",
"license": "ISC",
"workspaces": [
"./library",
+1 -1
View File
@@ -1,6 +1,6 @@
{
"name": "sdk",
"version": "1.15.0",
"version": "1.16.0",
"author": "",
"license": "ISC",
"description": "",
+1 -1
View File
@@ -1,7 +1,7 @@
[project]
name = "summary"
version = "1.15.0"
version = "1.16.0"
dependencies = [
"fastapi[standard]>=0.105.0",
"uvicorn>=0.24.0",
+3
View File
@@ -105,6 +105,9 @@ class Settings(BaseSettings):
# Speaker to user assignment
is_resolve_speaker_identities_enabled: bool = True
resolve_speaker_identities_default_overlap_threshold: float = 0.5
resolve_speaker_identities_enable_split_on_words: bool = True
resolve_speaker_identities_max_word_duration: float = 1 # seconds
# Webhook-related settings
webhook_max_retries: int = 2
+158 -21
View File
@@ -8,17 +8,18 @@ Multiple speakers can map to the same participant (e.g. two people sharing
one microphone). A participant with no matching speaker gets no assignment.
"""
import json
import logging
from collections import defaultdict
from dataclasses import dataclass, field
from dataclasses import asdict, dataclass, field, is_dataclass
from datetime import datetime
from typing import Any
logger = logging.getLogger(__name__)
from summary.core.config import get_settings
# Minimum fraction of a speaker's total duration that must overlap with a
# participant's VAD to accept the assignment.
DEFAULT_OVERLAP_THRESHOLD = 0.5
settings = get_settings()
logger = logging.getLogger(__name__)
@dataclass
@@ -74,7 +75,7 @@ class AssignmentResult:
f" ({item['speaker']})" if name_to_speaker_count[name] > 1 else ""
) # Add suffix only if there are multiple detected speakers per user
return {**item, "speaker": f"{name}{suffix}"}
return item
return {**item}
def _process_segment(
item: dict[str, Any], include_words: bool = False
@@ -138,6 +139,78 @@ def _overlap_duration(
return overlap
def _format_timelines_debug(
participant_timelines: dict[str, list[Interval]],
participant_names: dict[str, str],
speaker_timelines: dict[str, list[Interval]],
) -> str:
"""Render participant and speaker timelines side-by-side for debugging.
Each row is the slice between two consecutive interval boundaries
(drawn from both sides). A filled cell marks an active participant
(left block) or speaker (right block) during that slice, so vertical
alignment makes overlap visually obvious.
"""
participant_ids = sorted(participant_timelines.keys())
speaker_labels = sorted(speaker_timelines.keys())
if not participant_ids and not speaker_labels:
return "(no timelines)"
boundaries: set[float] = set()
for intervals in (*participant_timelines.values(), *speaker_timelines.values()):
for iv in intervals:
boundaries.add(iv.start)
boundaries.add(iv.end)
sorted_boundaries = sorted(boundaries)
if len(sorted_boundaries) < 2:
return "(no intervals)"
p_headers = [participant_names.get(pid, pid) for pid in participant_ids]
s_headers = list(speaker_labels)
p_widths = [max(len(h), 3) for h in p_headers]
s_widths = [max(len(h), 3) for h in s_headers]
def _active(intervals: list[Interval], lo: float, hi: float) -> bool:
mid = (lo + hi) / 2
return any(iv.start <= mid < iv.end for iv in intervals)
def _cells(
intervals_list: list[list[Interval]],
widths: list[int],
lo: float,
hi: float,
) -> str:
return " ".join(
("" * w if _active(iv, lo, hi) else "·" * w)
for iv, w in zip(intervals_list, widths, strict=True)
)
p_iv_list = [participant_timelines[pid] for pid in participant_ids]
s_iv_list = [speaker_timelines[sl] for sl in speaker_labels]
time_col = "[ start → end]"
p_hdr = (
" ".join(h.center(w) for h, w in zip(p_headers, p_widths, strict=True))
or "(none)"
)
s_hdr = (
" ".join(h.center(w) for h, w in zip(s_headers, s_widths, strict=True))
or "(none)"
)
sep = " || "
lines = [
f"{time_col} {p_hdr}{sep}{s_hdr}",
"-" * (len(time_col) + 2 + len(p_hdr) + len(sep) + len(s_hdr)),
]
for lo, hi in zip(sorted_boundaries, sorted_boundaries[1:], strict=False):
time_str = f"[{lo:8.2f}{hi:8.2f}]"
p_row = _cells(p_iv_list, p_widths, lo, hi) or " " * len(p_hdr)
s_row = _cells(s_iv_list, s_widths, lo, hi) or " " * len(s_hdr)
lines.append(f"{time_str} {p_row}{sep}{s_row}")
return "\n".join(lines)
def _build_participant_timelines(
metadata: dict[str, Any],
recording_start_datetime: datetime,
@@ -199,33 +272,77 @@ def _build_participant_timelines(
return intervals, participants_info
def _build_speaker_timelines(
transcription: Any,
) -> dict[str, list[Interval]]:
def _build_speaker_timelines(transcription: Any) -> dict[str, list[Interval]]:
"""Build interval timelines from WhisperX transcription segments."""
intervals: dict[str, list[Interval]] = {}
segments = transcription.segments if hasattr(transcription, "segments") else []
max_word_duration = settings.resolve_speaker_identities_max_word_duration
for segment in segments:
speaker = segment.get("speaker")
if speaker is None:
continue
intervals.setdefault(speaker, []).append(
Interval(segment["start"], segment["end"])
)
words = [
w
for w in segment.get("words", [])
if w.get("start") is not None and w.get("end") is not None
]
if not words:
intervals.setdefault(speaker, []).append(
Interval(segment["start"], segment["end"])
)
continue
start_time: float | None = segment["start"]
for word in words:
if start_time is None:
start_time = word["start"]
if not settings.resolve_speaker_identities_enable_split_on_words:
continue
if word["end"] - word["start"] > max_word_duration:
end_time = word["start"] + max_word_duration
if end_time > start_time:
intervals.setdefault(speaker, []).append(
Interval(start_time, end_time)
)
start_time = None
if start_time is not None:
last = words[-1]
end_time = min(last["end"], last["start"] + max_word_duration)
if end_time > start_time:
intervals.setdefault(speaker, []).append(Interval(start_time, end_time))
for speaker, speaker_intervals in intervals.items():
intervals[speaker] = _merge_intervals(speaker_intervals)
return intervals
def _json_default(obj: Any) -> Any:
"""Encode datetimes, dataclasses, and pydantic models for `json.dumps`.
Intended to be used for logging of `resolve_speaker_identities` (input
and computed variables)
"""
if isinstance(obj, datetime):
return obj.isoformat()
if is_dataclass(obj) and not isinstance(obj, type):
return asdict(obj)
if hasattr(obj, "segments") and hasattr(obj, "word_segments"):
return {"segments": obj.segments, "word_segments": obj.word_segments}
if hasattr(obj, "model_dump"):
return obj.model_dump(mode="json")
raise TypeError(f"Object of type {type(obj).__name__} is not JSON serializable")
def resolve_speaker_identities(
metadata: dict[str, Any],
transcription: Any,
recording_start_datetime: datetime,
recording_end_datetime: datetime,
overlap_threshold: float = DEFAULT_OVERLAP_THRESHOLD,
overlap_threshold: float = settings.resolve_speaker_identities_default_overlap_threshold, # noqa: E501
) -> AssignmentResult:
"""Match WhisperX speaker labels to participants.
@@ -246,12 +363,6 @@ def resolve_speaker_identities(
)
speaker_timelines = _build_speaker_timelines(transcription)
logger.debug(
"Assignment inputs: %d participants, %d speakers",
len(participant_timelines),
len(speaker_timelines),
)
result = AssignmentResult()
for speaker, speaker_intervals in speaker_timelines.items():
@@ -294,4 +405,30 @@ def resolve_speaker_identities(
overlap_threshold,
)
logger.debug(
json.dumps(
{
"input": {
"recording_start_datetime": recording_start_datetime.isoformat(),
"recording_end_datetime": recording_end_datetime.isoformat(),
"metadata": metadata,
"transcription": transcription,
},
"computed": {
"speaker_timelines": speaker_timelines,
"participant_timelines": participant_timelines,
"result": result,
},
},
default=_json_default,
indent=2,
ensure_ascii=False,
),
)
logger.debug(
_format_timelines_debug(
participant_timelines, participant_names, speaker_timelines
),
)
return result
+180
View File
@@ -4,10 +4,12 @@ import math
from dataclasses import dataclass, field
from datetime import datetime
from summary.core import user_assign
from summary.core.user_assign import (
AssignmentResult,
Interval,
SpeakerAssignment,
_build_speaker_timelines,
_merge_intervals,
_overlap_duration,
_total_duration,
@@ -73,12 +75,22 @@ DIARIZATION_SINGLE_SPEAKER = FakeTranscription(
"end": 3.545,
"text": " The stale smell.",
"speaker": "SPEAKER_00",
"words": [
{"word": "The", "start": 1.363, "end": 1.8},
{"word": "stale", "start": 1.8, "end": 2.7},
{"word": "smell.", "start": 2.7, "end": 3.545},
],
},
{
"start": 4.466,
"end": 6.247,
"text": "It takes heat.",
"speaker": "SPEAKER_00",
"words": [
{"word": "It", "start": 4.466, "end": 4.7},
{"word": "takes", "start": 4.7, "end": 5.5},
{"word": "heat.", "start": 5.5, "end": 6.247},
],
},
],
)
@@ -165,6 +177,174 @@ class TestTotalDuration:
assert math.isclose(_total_duration([]), 0.0)
class TestBuildSpeakerTimelines:
"""Tests for _build_speaker_timelines."""
def test_segment_without_words_falls_back_to_segment_bounds(self):
"""Segments missing a `words` key use the segment start/end as one interval."""
transcription = FakeTranscription(
segments=[{"start": 1.5, "end": 3.5, "speaker": "SPEAKER_00"}],
)
result = _build_speaker_timelines(transcription)
assert result == {"SPEAKER_00": [Interval(1.5, 3.5)]}
def test_segment_with_only_none_word_timestamps_falls_back(self):
"""If every word has None start/end, fall back to segment bounds."""
transcription = FakeTranscription(
segments=[
{
"start": 1.0,
"end": 4.0,
"speaker": "SPEAKER_00",
"words": [
{"word": "hi", "start": None, "end": None},
{"word": "there", "start": None, "end": None},
],
},
],
)
result = _build_speaker_timelines(transcription)
assert result == {"SPEAKER_00": [Interval(1.0, 4.0)]}
def test_short_words_only_uses_segment_start_and_last_word_end(self):
"""With no overly long words, the interval runs segment start to end."""
transcription = FakeTranscription(
segments=[
{
"start": 1.0,
"end": 5.0,
"speaker": "SPEAKER_00",
"words": [
{"word": "a", "start": 1.0, "end": 1.3},
{"word": "b", "start": 1.4, "end": 1.7},
{"word": "c", "start": 1.8, "end": 2.1},
],
},
],
)
result = _build_speaker_timelines(transcription)
# Tail: min(2.1, 1.8 + 1.0) = 2.1
assert result == {"SPEAKER_00": [Interval(1.0, 2.1)]}
def test_long_word_caps_interval_at_max_duration(self):
"""A word longer than the max-word-duration cap truncates the segment."""
max_word_duration = (
user_assign.settings.resolve_speaker_identities_max_word_duration
)
transcription = FakeTranscription(
segments=[
{
"start": 0.0,
"end": max_word_duration + 7,
"speaker": "SPEAKER_00",
"words": [
{
"word": "pause",
"start": 0.0,
"end": max_word_duration + 7,
},
],
},
],
)
result = _build_speaker_timelines(transcription)
assert result == {"SPEAKER_00": [Interval(0.0, max_word_duration)]}
def test_long_word_in_middle_splits_segment(self):
"""Short words around a long word produce two intervals (before-cap + after)."""
transcription = FakeTranscription(
segments=[
{
"start": 0.0,
"end": 20.0,
"speaker": "SPEAKER_00",
"words": [
{"word": "a", "start": 0.0, "end": 0.5},
{"word": "long", "start": 1.0, "end": 15.0},
{"word": "z", "start": 18.0, "end": 18.4},
],
},
],
)
result = _build_speaker_timelines(transcription)
# First emit: (0.0, 1.0 + 1.0). Then start_time resets, picks up at "z" (18.0).
# Tail: min(18.4, 18.0 + 1.0) = 18.4. So second interval is (18.0, 18.4).
assert result == {
"SPEAKER_00": [Interval(0.0, 2.0), Interval(18.0, 18.4)],
}
def test_tail_word_is_capped_at_max_duration(self):
"""The trailing word's end is capped at word.start + max_word_duration."""
transcription = FakeTranscription(
segments=[
{
"start": 0.0,
"end": 50.0,
"speaker": "SPEAKER_00",
"words": [
{"word": "a", "start": 0.0, "end": 0.4},
# Last word ends inside the cap, so the cap doesn't apply.
{"word": "b", "start": 1.0, "end": 1.5},
],
},
],
)
result = _build_speaker_timelines(transcription)
# Tail: min(1.5, 1.0 + 1.0) = 1.5
assert result == {"SPEAKER_00": [Interval(0.0, 1.5)]}
def test_split_on_words_disabled_keeps_segment_as_one_interval(self, monkeypatch):
"""With splitting disabled, long words don't split the interval."""
monkeypatch.setattr(
user_assign,
"settings",
user_assign.settings.model_copy(
update={"resolve_speaker_identities_enable_split_on_words": False},
),
)
transcription = FakeTranscription(
segments=[
{
"start": 0.0,
"end": 20.0,
"speaker": "SPEAKER_00",
"words": [
{"word": "a", "start": 0.0, "end": 0.5},
{"word": "long", "start": 1.0, "end": 15.0},
{"word": "z", "start": 18.0, "end": 18.4},
],
},
],
)
result = _build_speaker_timelines(transcription)
# No mid-segment split; tail caps at min(18.4, 18.0 + 1.0) = 18.4.
assert result == {"SPEAKER_00": [Interval(0.0, 18.4)]}
def test_multiple_speakers_keep_separate_timelines(self):
"""Segments from different speakers populate independent timeline entries."""
transcription = FakeTranscription(
segments=[
{
"start": 0.0,
"end": 1.0,
"speaker": "SPEAKER_00",
"words": [{"word": "hi", "start": 0.0, "end": 0.5}],
},
{
"start": 2.0,
"end": 3.0,
"speaker": "SPEAKER_01",
"words": [{"word": "yo", "start": 2.0, "end": 2.5}],
},
],
)
result = _build_speaker_timelines(transcription)
assert result == {
"SPEAKER_00": [Interval(0.0, 0.5)],
"SPEAKER_01": [Interval(2.0, 2.5)],
}
class TestResolveSpeakerIdentities:
"""Tests for resolve_speaker_identities."""