mirror of
https://github.com/suitenumerique/meet.git
synced 2026-07-28 04:39:16 +00:00
Compare commits
8 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
 lebaudantoine
|
35951ba2a6 | ||
|
lebaudantoine
|
72184e1370 | ||
|
leo
|
1b4a8fbac2 | ||
|
lebaudantoine
|
1e2fad5444 | ||
|
leo
|
96f97ed2d0 | ||
|
lebaudantoine
|
02d16cb55c | ||
|
lebaudantoine
|
7268ff6777 | ||
|
lebaudantoine
|
cca5bc2186 |
+4
-3
@@ -8,6 +8,8 @@ and this project adheres to
|
||||
|
||||
## [Unreleased]
|
||||
|
||||
## [1.16.0] - 2026-05-13
|
||||
|
||||
### Added
|
||||
|
||||
- 🔒️(backend) add validation of Room.configuration
|
||||
@@ -19,10 +21,8 @@ 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 +32,7 @@ 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
|
||||
|
||||
|
||||
@@ -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
@@ -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:
|
||||
|
||||
@@ -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=
|
||||
|
||||
Generated
+12
-250
@@ -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",
|
||||
|
||||
@@ -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",
|
||||
|
||||
@@ -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 ----
|
||||
|
||||
@@ -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}")
|
||||
|
||||
|
||||
@@ -1,7 +1,7 @@
|
||||
|
||||
[project]
|
||||
name = "agents"
|
||||
version = "1.15.0"
|
||||
version = "1.16.0"
|
||||
requires-python = ">=3.12"
|
||||
dependencies = [
|
||||
"livekit-agents==1.4.5",
|
||||
|
||||
Generated
+15
-15
@@ -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]]
|
||||
|
||||
@@ -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)
|
||||
@@ -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",
|
||||
|
||||
Generated
+1
-1
@@ -1173,7 +1173,7 @@ wheels = [
|
||||
|
||||
[[package]]
|
||||
name = "meet"
|
||||
version = "1.15.0"
|
||||
version = "1.16.0"
|
||||
source = { editable = "." }
|
||||
dependencies = [
|
||||
{ name = "aiohttp" },
|
||||
|
||||
Generated
+421
-462
File diff suppressed because it is too large
Load Diff
@@ -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",
|
||||
|
||||
@@ -12,7 +12,7 @@ const controlBarRegion = cva({
|
||||
variants: {
|
||||
mobile: {
|
||||
true: {
|
||||
justifyContent: 'space-between',
|
||||
justifyContent: 'center',
|
||||
width: '330px',
|
||||
},
|
||||
},
|
||||
|
||||
@@ -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,
|
||||
},
|
||||
]
|
||||
})
|
||||
}
|
||||
|
||||
|
||||
Generated
+5
-5
@@ -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"
|
||||
|
||||
@@ -1,6 +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": {
|
||||
|
||||
Generated
+2
-2
@@ -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",
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
{
|
||||
"name": "sdk",
|
||||
"version": "1.15.0",
|
||||
"version": "1.16.0",
|
||||
"author": "",
|
||||
"license": "ISC",
|
||||
"description": "",
|
||||
|
||||
@@ -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",
|
||||
|
||||
@@ -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
|
||||
|
||||
@@ -14,11 +14,11 @@ from dataclasses import dataclass, field
|
||||
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 +74,7 @@ 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 +138,78 @@ 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,
|
||||
@@ -199,24 +271,50 @@ 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
|
||||
|
||||
|
||||
@@ -225,7 +323,7 @@ def resolve_speaker_identities(
|
||||
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.
|
||||
|
||||
@@ -247,9 +345,14 @@ def resolve_speaker_identities(
|
||||
speaker_timelines = _build_speaker_timelines(transcription)
|
||||
|
||||
logger.debug(
|
||||
"Assignment inputs: %d participants, %d speakers",
|
||||
"Assignment inputs: %d participants, %d speakers\n%s\n%s\n%s",
|
||||
len(participant_timelines),
|
||||
len(speaker_timelines),
|
||||
participant_timelines,
|
||||
speaker_timelines,
|
||||
_format_timelines_debug(
|
||||
participant_timelines, participant_names, speaker_timelines
|
||||
),
|
||||
)
|
||||
|
||||
result = AssignmentResult()
|
||||
|
||||
@@ -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."""
|
||||
|
||||
|
||||
Reference in New Issue
Block a user