lucasw89/whisper_web
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
145 Commits 1 Branch 0 Tags
main
Commit Graph
145 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Lucas WintherandClaude Opus 5.5 80f8004963 Close moot backlog items; refresh verified commands and suite timing 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 14:37:24 +01:00
Lucas WintherandClaude Opus 5.5 978743a36e Record that batched Opus-MT is still broken on Transformers.js 4.3.2 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 14:36:51 +01:00
Lucas WintherandClaude Opus 5.5 dd9dfc628b Detect the language once for auto-detect file passes on the chunked path (D-053) 
Transformers.js has no Whisper language detection and silently decoded
Danish as English whenever "Skip silence" was off or unavailable.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 14:35:09 +01:00
Lucas WintherandClaude Opus 5.5 3583d60847 Reject task=translate on the server for models that cannot translate (D-044) 
The UI already hid it; a direct API call to POST /api/transcribe or the live
WebSocket could still get the source language back labelled as a translation.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 14:33:08 +01:00
Lucas WintherandClaude Opus 5.5 96d100bd45 Turn "Skip silence" on by default (opt-out, D-052) 
Owner's decision: the long-form pipeline's gain on the long real recording
outweighs the spec's strict every-clip rollout rule. A stored "off" still wins.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:56:57 +01:00
Lucas WintherandClaude Opus 5.5 0207ff5a3d Spec 001 implemented; remove the coordination plan 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:49:43 +01:00
Lucas WintherandClaude Opus 5.5 0632f7843b Renumber spec 001 decision placeholders to D-040–D-051; record the Danish publication 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:47:27 +01:00
Lucas Winther 30969e86b7 Merge feat/parakeet-engine: Parakeet TDT 0.6B v3 English engine (spec 001 workstream 7) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>

# Conflicts:
#	AGENTS.md
#	bun.lock
#	docs/Decisions.md
#	src/lib/models.ts
#	src/worker/inference.worker.ts
#	src/worker/protocol.ts
#	tests/models.test.ts
 2026-10-30 01:46:51 +01:00
Lucas Winther 9c89ab047f Merge feat/longform-pipeline: VAD long-form pipeline, opt-in (spec 001 workstream 3) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>

# Conflicts:
#	docs/Architecture.md
#	docs/Decisions.md
#	src/App.tsx
#	src/index.css
#	src/lib/engine.ts
#	src/worker/inference.worker.ts
#	src/worker/protocol.ts
#	tests/engine.test.ts
#	tests/live-real-model.test.ts
#	tests/offline.md
#	tests/protocol.test.ts
 2026-10-30 01:39:17 +01:00
Lucas Winther 2c87471fdc Merge feat/vocabulary-hint: names-and-terms decoder prompt (spec 001 workstream 2) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>

# Conflicts:
#	docs/Architecture.md
#	docs/Decisions.md
#	docs/WorkBreakdown.md
#	server-py/sidecar.py
#	src/App.tsx
#	src/server/live.ts
#	src/server/transcribe.ts
#	tests/fixtures/fake_sidecar.py
#	tests/transcribe.test.ts
 2026-10-30 01:35:28 +01:00
Lucas WintherandClaude Opus 5.5 5f335c3e41 Publish only whisper-small-danish under lucasw89; drop the turbo Danish entries 
The owner chose to publish just the small fine-tune: the turbo fine-tunes
showed no clear gain over stock turbo on da_festuge (D-WS6a).

Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:32:30 +01:00
Lucas Winther aa7ae93c0c Merge feat/server-batched-vad: faster-whisper batched + VAD (spec 001 workstream 5) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>

# Conflicts:
#	docs/Decisions.md
 2026-10-30 01:27:55 +01:00
Lucas Winther 439d099279 Merge feat/danish-models: Danish fine-tune catalog entries (spec 001 workstream 6) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>

# Conflicts:
#	docs/Decisions.md
#	docs/WorkBreakdown.md
 2026-10-30 01:27:45 +01:00
Lucas Winther c94689e2e0 Merge feat/translate: canTranslate + Opus-MT transcript translation (spec 001 workstream 4) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>

# Conflicts:
#	docs/Decisions.md
 2026-10-30 01:15:05 +01:00
Lucas WintherandClaude Opus 5.5 316f64609b Merge chore/transformers-4.3: Transformers.js 4.3.2 (spec 001 workstream 1) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:14:22 +01:00
Lucas WintherandClaude Opus 5.5 498e361b8a Remove the workstream 6 plan now that the Danish models work is done 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:13:27 +01:00
Lucas WintherandClaude Opus 5.5 19228297f1 Record the Danish conversions' measurements, dtype choice and pending upload in D-WS6a 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:13:27 +01:00
Lucas WintherandClaude Opus 5.5 951aad8f55 Default the Danish conversions to q4 on WASM, as measured on the converted files 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:12:28 +01:00
Lucas WintherandClaude Opus 5.5 7b65f1f727 Remove the finished Transformers.js 4.3.2 upgrade plan 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 01:04:18 +01:00
Lucas WintherandClaude Opus 5.5 cc8e031d6b Run WASM sessions at the default graph optimization level now that ort-web 1.31 loads the 8-bit decoders 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:45:46 +01:00
Lucas WintherandClaude Opus 5.5 cb80ab5f9d Remove plan 001-3; spec 001 and D-WS3a/D-WS3b are the record 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:43:51 +01:00
Lucas WintherandClaude Opus 5.5 215035e383 Assert whole-file WER per clip in the real-model test (plan 001-3 Tasks 8-9 done) 
Chunked file path and long-form (explicit and auto language) held to the
per-clip bars; full real-model test green under the heavy-job lock.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:43:51 +01:00
Lucas WintherandClaude Opus 5.5 7fb8dde423 Remove the finished server batched-VAD plan; spec 001 and D-WS5a are the record 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:24:06 +01:00
Lucas WintherandClaude Opus 5.5 07753a496b Tick the finished server batched-VAD plan 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:24:06 +01:00
Lucas WintherandClaude Opus 5.5 104c1c4e4d Record the batched-VAD measurements and check heartbeat, bursts and cancel against the real sidecar 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:23:55 +01:00
Lucas WintherandClaude Opus 5.5 ab45e686a4 Record the long-form pipeline measurements, rollout and VAD model (D-WS3a, D-WS3b) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:23:21 +01:00
Lucas WintherandClaude Opus 5.5 b88f406ac6 Remove the workstream 4 plan now that it is complete 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:19:24 +01:00
Lucas WintherandClaude Opus 5.5 160298197e Reuse the language-detection encoder pass in long-form decodes 
Whisper's generate() keeps only forward_params from its kwargs, and that
list omits encoder_outputs, so the precomputed encoder output was dropped
and auto-detect cost three encoder passes for a one-chunk clip. The
real-model test now pins one encoder pass per decoded chunk.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:19:08 +01:00
Lucas WintherandClaude Opus 5.5 c0528af082 Wrap transcript card actions under the title on narrow screens 
Polish pass on the translate panel: the four export buttons overflowed
both transcript cards at 390 px. Also record the headless-Chrome
translation run and a re-measured bench in D-WS4b.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:19:04 +01:00
Lucas WintherandClaude Opus 5.5 6b338a2c58 Remove the finished workstream 2 plan (vocabulary hint) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:13:42 +01:00
Lucas WintherandClaude Opus 5.5 4e65e322ce Tick plan 001-2 Task 7: real-model file runs green under the lock (7 pass, 920 s) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:13:42 +01:00
Lucas WintherandClaude Opus 5.5 1c1806840e Tune the server engine's Silero VAD so batched file transcription stops dropping quiet speech 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-30 00:07:05 +01:00
Lucas WintherandClaude Opus 5.5 b66ba7e577 Remove the workstream 7 plan now that the Parakeet engine path is done 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:52:39 +01:00
Lucas WintherandClaude Opus 5.5 f540f8952d Keep the noticed-not-fixed list contiguous 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:52:35 +01:00
Lucas WintherandClaude Opus 5.5 378aff6a7b Record D-WS2a: hint measurements, the 120-token truncation rule, no faster-whisper hotwords 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:52:29 +01:00
Lucas WintherandClaude Opus 5.5 fdc9b500c0 Check Parakeet chunks export to .srt and timestamped text, and record the headless Chrome run 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:52:05 +01:00
Lucas WintherandClaude Opus 5.5 edb9b332b9 Cap the hint prompt at 120 tokens: a full 223-token prompt garbles small models 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:51:47 +01:00
Lucas WintherandClaude Opus 5.5 9b37c7d26c Tick the server batched-VAD plan up to the committed Bun and sidecar work 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:48:14 +01:00
Lucas WintherandClaude Opus 5.5 28201bcd17 Reconcile plan 001-2 with the work done before the pause 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:34:36 +01:00
Lucas WintherandClaude Opus 5.5 7ff23628e3 Tick the Parakeet plan steps already landed before the pause 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:34:28 +01:00
Lucas WintherandClaude Opus 5.5 8ac81baf54 Tick plan 001-3 steps already done before the pause 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:34:13 +01:00
Lucas WintherandClaude Opus 5.5 dc15623f62 WIP: uncommitted work at pause (session limit) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:03:27 +01:00
Lucas WintherandClaude Opus 5.5 826bcd2e49 WIP: uncommitted work at pause (session limit) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:03:27 +01:00
Lucas WintherandClaude Opus 5.5 ec7ed36426 WIP: uncommitted work at pause (session limit) 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 23:03:27 +01:00
Lucas WintherandClaude Opus 5.5 9f1d58c0b7 Keep max_new_tokens: 440 on Transformers.js 4.3.2 and pin the seek-loop condition it avoids 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 22:59:35 +01:00
Lucas WintherandClaude Opus 5.5 7dc77978a9 Classify a silent translate-worker crash as out of memory 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 22:51:47 +01:00
Lucas WintherandClaude Opus 5.5 9cd92b4920 Record D-WS7a: Parakeet runtime decision, engine path and CPU measurements vs Large v3 Turbo 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 22:32:22 +01:00
Lucas WintherandClaude Opus 5.5 cd9aa48b83 Record what the Transformers.js 4.3.2 upgrade changes on the WebGPU side and what needs a browser check 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 22:30:29 +01:00
Lucas WintherandClaude Opus 5.5 8e0ecdaf85 Add the synthetic meeting fixture and long-form measurement script 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 22:28:48 +01:00
Lucas WintherandClaude Opus 5.5 a569b4003c Tick completed workstream 4 plan steps 
Co-Authored-By: Claude Opus 5.5 <[email protected]>
 2026-10-29 22:11:56 +01:00