Repository navigation
Utterform 0.13.0 Voice messages and recordings - #51
Merged
Merged
Conversation
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014LfoKeUu4n9ax82oQFfKHo
Voice messages and recordings are decoded to 16 kHz mono with Symphonia and a bundled libopus, then take the recording path: transcription, history, clipboard and text actions. Opened from a header button or O, dropped on the window, passed as --file or by Open with. Audio longer than twelve minutes is uploaded to GPT Transcribe in parts cut at pauses. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014LfoKeUu4n9ax82oQFfKHo
…-audio-files # Conflicts: # src/lib/api.ts # tests/ui.e2e.ts
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_014LfoKeUu4n9ax82oQFfKHo
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Stage 0.13.0 of the 1.0 roadmap: transcribe voice messages and recordings.
O, drop on the window ("Drop to transcribe" with action and engine),utterform --file <path>, and "Open with" (Linux desktop entry, WindowsOpenWithProgidsin the NSIS hook, macOSCFBundleDocumentTypeswith rank Alternate — none becomes the default player). Files from outside queue and are done one after another.audio_file.rs— Symphonia (Ogg, WebM/MKV, MP4/M4A, MP3, WAV, FLAC, AIFF, CAF) plus libopus bundled viasymphonia-adapter-libopusfor WhatsApp/Telegram Opus; streamed to a temporary 16 kHz mono WAV, then the normalprocess_recordingpath withOrigin::File(history, clipboard, action, never the cursor; no kept copy on failure). Two-hour limit.file.requested/started/decoded/failed,transcription.split/part_failed,recording.completed mode=file; no file names or paths.Validation:
npm run check,npm test(153),npm run build, Playwright (36),cargo fmt --check,cargo clippy -D warnings,dbus-run-session -- cargo test(209). Live on asus16: invented WhatsApp-style Opus message via GPT Transcribe (1.6 s) and the built app with local Whisper (4 s), a 13-minute file in two parts, second-instance hand-off with a relative path,install-linux.shdesktop entry validated. Open for 1.0 (in TESTING.md): Windows/macOS "Open with" and drag and drop in a real window; libopus on the Windows/macOS runners.🤖 Generated with Claude Code
https://claude.ai/code/session_014LfoKeUu4n9ax82oQFfKHo