Repository navigation
fix(stt): accept all profile languages on transcribe/retranscribe - #1198
Open
vincentfranstyo wants to merge 1 commit into
Open
vincentfranstyo wants to merge 1 commit into
vincentfranstyo wants to merge 1 commit into
Conversation
TranscriptionRequest and CaptureRetranscribeRequest only allowed the original ten Qwen languages, so codes like pl returned 422 even though Whisper handles them. Use the same 23-language pattern as profiles.
doublesharp
added a commit
to doublesharp/voicebox-ds
that referenced
this pull request
Oct 29, 2026
Use the shared QueryClient for offline requests, preserve custom server URLs through startup and migration, retain capture decode diagnostics, preserve refinement language, accept supported STT languages, and explain LuxTTS short-input errors. Adapt upstream PRs jamiepine#1208, jamiepine#1204, jamiepine#1219, jamiepine#1218, jamiepine#1198, jamiepine#1199, jamiepine#1220 and jamiepine#1180; retain remote reconnect refreshes and correctly migrate the legacy bundled port. Wait for the preview output context to resume before playing. Cancel superseded starts and release preview nodes/object URLs on replacement, completion, failure and unmount. Add CI regressions for the upstream integration and preview lifecycle. Validation: 292 frontend tests on this candidate; 414 frontend tests with the working UI changes; 53 backend tests; app/web/Tauri frontend typechecks and both frontend builds. Independent review accepted the playback change. Native first-click/autoplay, acoustic output, packaging and model quality remain unverified. Co-authored-by: vincentfranstyo <66989751+vincentfranstyo@users.noreply.github.com> Co-authored-by: maxazure <405711+maxazure@users.noreply.github.com> Co-authored-by: FelixKruger <262215022+FelixKruger@users.noreply.github.com> Co-authored-by: Tony25032 <141424750+Tony25032@users.noreply.github.com> Co-authored-by: mvoloch <274143174+mvoloch@users.noreply.github.com> Co-authored-by: Lvigentini <11659311+Lvigentini@users.noreply.github.com>
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
POST /transcribeandPOST /captures/{id}/retranscribereject Polish (and every other language added after the original ten) with a 422, even though profiles and/speakaccept them and Whisper transcribes them fine:TranscriptionRequest.languageandCaptureRetranscribeRequest.languagestill use the old^(en|zh|ja|ko|de|fr|ru|pt|es|it)$pattern.Fix
Both fields now use the same 23-language pattern as
VoiceProfileCreate/GenerationRequest. No other code changes; the Whisper backends already pass the code straight through.Testing
backend/tests/test_stt_language.py: checks both STT models share the profile pattern, acceptpl, still default toNone, and still reject unknown codes. Fails onmain, passes with this change.ruff check/ruff format --checkclean on the new test.Fixes #1095