We should be able to use other transcription services,
those usually relie on response_format="diarized_json"
to produce what we need.
Note that as part of this change, we stop using
openai library for making this call to avoid
casting the result to a payload that doesn't
contain the elements we used to rely on.
(setting this specific format auto cast the
results in openai lib). We keep the old
result class used.
When going through AlbertAPI, timestamp are not provided at the word level.
This adds default values so that the summary external contract stays the same,
while giving us compatibility wisht AlbertAPI.
Updates the help chart to be able to run a list of CronJobs.
By defaults it runs the clean_pending_files
and purge_deleted_files commands during the night.
Since we have added custom backgrounds, they could be
soft deleted but were never really deleted.
This commits introduces a command to actually delete
this files.
In introduces also a command to purge pending upload
.
Add a footer to transcription outputs linking to an external satisfaction
survey. The survey URL is built from TRANSCRIPTION_SATISFACTION_FORM_BASE_URL.
When TRANSCRIPTION_SATISFACTION_FORM_BASE_URL is unset or None, the
footer is omitted.
When joining a meeting that already has many participants, new
participants are now muted by default.
This is an empirical change, not directly requested by users but
informed by user experience: a lot of people joining large meetings
arrive with their microphone and/or camera open, which can be
painful for the host, who otherwise has to mute everyone or ask
everyone to mute. Many of these participants are inattentive but
still noisy.
If you are joining a room that is already at a certain size, you are
probably not the one expected to speak; presenters typically join
among the very first participants.
In particularly large rooms, with more than 5 or 10 participants,
the entry notification sound can quickly feel spammy.
A configuration to disable it exists, but new users do not discover
it easily. The sound notification is most useful while waiting for
the first few participants to arrive; once a few people are in, the
meeting usually starts and the sound becomes more disruptive than
helpful.
The visual notification remains in place, so users are still aware
when newcomers join. Roll this out and check whether users find the
change helpful.
Required by some self-hosters who want to present a link to
participants that are guests only, without access to their SSO. For
example, a meeting between a citizen and a public servant where the
guest is known not to have an account.
Also useful when rendering the join page inside an iframe.
Several clients ran into issues with the silent login, leading to
redirections that were not appropriate for their use case.
Offer a URL parameter to control this behavior and disable silent
login when needed.
The noise-suppression processor is now imported on demand with a
dynamic import() instead of a top-level static import, so it lands in
its own code-split chunk rather than the main bundle.
Why this is necessary:
* @libreaudio/la-call is fully self-contained. It inlines *everything*
as JavaScript: two WASM binaries (SIMD and non-SIMD variants), the
Emscripten glue, the worklet processor source, and the noise model
plus its weights — the model is compiled into the .wasm, so we ship
effectively two copies of it.
* The WASM is inlined as numeric array literals (new Uint8Array([...])),
the least compact representation possible (~2-4 source chars/byte) and
not meaningfully minifiable. The result is a large module that also
can't be stream-compiled the way an external .wasm asset would be.
* A static import would pull all of that into the initial bundle,
inflating critical-path download size and lengthening build time
(parsing/minifying the big literal) — for a feature that's only used
when the user actually turns on noise suppression.
* Dynamic import() isolates the whole payload in a separate,
content-hashed, browser-cached chunk. The cost (chunk fetch + WASM
instantiation) becomes a one-time hit deferred to first activation,
and is fully off the page's initial load path.
Notes:
* The library self-bundles its AudioWorklet at runtime from a Blob URL
and inlines the WASM, so it needs no build-tool asset plumbing
(no ?worker&url / ?url, no Vite asset config). The dynamic import is
therefore the only splitting mechanism required.
* Audio processing still runs off the main thread on the AudioWorklet
path (desktop/most browsers); only the Android ScriptProcessor
fallback runs on the main thread. Import style does not affect this.
* To hide the first-activation latency, the chunk can be prefetched
(e.g. import() on idle or <link rel="modulepreload">) so it's warm
before the user enables suppression.
Replace the basic RNN noise processor with a more advanced audio
pipeline powered by Big Blue Better Audio (BBBA), a project
supported by the Prototype Fund.
The new WASM-based pipeline introduced by @falkTX and
@trummerschlunk adds:
- voice isolation
- high-pass filtering
- spectral balancing
- multiband compression
- low-pass filtering
This significantly improves overall audio quality and speech
clarity.
More information about the pipeline is available on the
trummerschlunk/BigBlueBetterAudio repo.
Voice isolation still relies on RNNNoise, a widely used deep
learning-based denoising model. Audio quality could be further
improved in the future if DeepFilterNet becomes usable directly
in the browser.
Their code has been released in an NPM package under GPL license.
A previous CSP change suggested by CodeRabbit was not properly
tested and broke inline styling as well as the loading of the
ProConnect image.
Adjust the CSP directives to allow these resources again.
The CSP was missing a default-src directive (flagged by CodeRabbit).
Without it, styles, images, fonts, and media are completely
unrestricted, which undercuts the otherwise strict policy.
Set default-src to 'self' and add an explicit style-src directive
since we rely on inline styles (e.g. #close-msg and the view
toggling use inline style attributes), so style-src needs to allow
'self' and 'unsafe-inline'.
Mount the nginx configuration used by the frontend image as a volume
so the default one can be overridden by a custom configuration at
deployment time.
Based on suggestions from @rouja to help parametrize the
configuration at deployment time.
Add the possibility to display a feedback form link in the footer to
help collect feedback from the first users of the plugin.
Once the plugin is in production at scale, this will be replaced
with a link to the support page.
Offer the possibility to append a query parameter at the end of the
meeting URL to flag links created from the Outlook plugin, inspired
by Zoom.
The behavior is toggled via an environment variable.
From the task panel, change the "add" button to "remove" when a
meeting link is already present.
This will be updated to "update" once the external API supports
updating an existing room.
Detect whether a meeting link is already present in the calendar
event item or its body, and prevent generating a new link if one is
found.
Detection is based on the presence of the app URL in the text. This
is imperfect but covers the most naive scenarios.
Refactor the plugin to support internationalization and ship it in
three languages.
Internationalizing requires updating the manifest, which involves
touching the Helm chart and releasing a new version. Ship a beta
version of the plugin once the i18n work is done to bundle these
changes together.
In the ribbon, the "create a meeting link" action was displayed with
the generic add icon, which is fitting in the context of a menu but
not explicit enough on its own.
Following feedback from the social ministries, replace the add icon
with the app logo to make it clear that this button creates a
videoconference link.
Instead of appending the meeting link at the end of the email, insert
it where the user's cursor is. This ensures the link is not placed
after the email signature, or below the quoted thread in a reply.
Refactor the message builder so that rendered messages are displayed
in a nicer way. Based on user feedback from the social ministries.
Go for a hybrid approach: on the desktop client, insert plain text so
it picks up the default font configured by the client theme; on the
web client, insert HTML, as the editor there is rendered using HTML.
The dialog window does not close itself on Outlook Desktop, which
opens links in a webview that behaves like a different browser.
Calling window.close() fails because, for security reasons, JS is
prevented from closing the window. The browser considers that the JS
script is not the one that opened the dialog.
The dialog also loses its reference to the opener due to the
redirection, so there is no way to message the parent to trigger a
close.
Keeping the dialog reference on the parent side does not help either:
since the opener is lost, the parent cannot call close on the dialog.
After investigation, go with a temporary solution that shows an
explicit message hinting the user to close the dialog manually if it
does not close automatically.
Restrict arm64 builds to release tags only, instead of running them on
every build.
The arm64 pipeline was significantly slowing down CI with little
practical usage, degrading developer experience for regular workflows.
Keep arm64 builds for release validation, and handle occasional failures
manually if needed.
The LiveKit integration was still using RoomInputOptions and
RoomOutputOptions, which emit deprecation warnings.
Update the implementation to use the unified RoomOptions API.
Pin urllib3 to >=2.7.0 via uv constraint-dependencies to fix a moderate
severity decompression DoS vulnerability. Affected versions (2.6.0 to
<2.7.0) could fully decode a compressed response body in a single
operation, leading to excessive CPU and memory consumption.
Pin idna to >=3.15 via uv constraint-dependencies to fix a moderate
severity (CVSS 5.3) information disclosure vulnerability. The flaw is
network-exploitable with no authentication required.
snyk-bot