tomatocream
843ec534d1
fix: handle processor.decode returning a list of strings
...
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-05-29 03:00:09 +08:00
tomatocream
cf18335235
fix: simplify audio callback, use deque for pre-roll, add worker timeout warning
...
- Remove frame_buf accumulation: blocksize=FRAME_SIZE guarantees indata is
exactly FRAME_SIZE samples, so buffering was unnecessary. Use indata[:, 0].copy()
to avoid stale references from sounddevice's buffer reuse.
- Replace pre_roll list with collections.deque(maxlen=PRE_ROLL_FRAMES) to
eliminate manual bounds-checking (pop(0)) on every frame.
- Warn to stderr if the transcription worker thread outlives its 30s join timeout.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-29 02:48:51 +08:00
tomatocream
747a4772b6
feat: implement live streaming transcription with VAD
...
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-29 02:46:13 +08:00
tomatocream
d62fcdd1cd
feat: add silence calibration and VAD state machine
...
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-29 02:45:09 +08:00
tomatocream
4605be5bc9
refactor: switch to argparse, add --stream and --lang flags
...
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com >
2026-05-29 02:43:47 +08:00
tomatocream
6bff2875c5
Add implementation plan for live streaming transcription
...
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-05-29 02:42:00 +08:00
tomatocream
e0911653fe
Add design spec for live streaming microphone transcription
...
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com >
2026-05-29 02:38:05 +08:00
tomatocream
c055a8ffb9
Replace flake.nix with shell.nix for simpler NixOS dev environment
2026-05-26 01:59:15 +08:00
tomatocream
55a51a7668
Add flake.nix with portaudio + CUDA, microphone support in transcribe.py
2026-05-26 01:55:54 +08:00
tomatocream
8b88489a53
Simplify to audio file input (mic requires PortAudio on NixOS)
2026-05-26 01:49:52 +08:00
tomatocream
14abcb89f2
Add accelerate dependency
2026-05-26 01:38:10 +08:00
tomatocream
82fe21fe41
Add Cohere Transcribe demo with uv + Python 3.14
2026-05-26 01:35:10 +08:00
tomatocream