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 <noreply@anthropic.com>

2026-10-30 00:19:08 +01:00

..

components

Run browser file transcription through the long-form pipeline when asked and the VAD model is available

2026-10-29 22:02:47 +01:00

lib

Run browser file transcription through the long-form pipeline when asked and the VAD model is available

2026-10-29 22:02:47 +01:00

pwa

subpath bugfixes

2026-07-10 10:20:23 +02:00

server

Usability and test infrastructure updates.

2026-07-29 09:20:43 +02:00

worker

Reuse the language-detection encoder pass in long-form decodes

2026-10-30 00:19:08 +01:00

App.tsx

Run browser file transcription through the long-form pipeline when asked and the VAD model is available

2026-10-29 22:02:47 +01:00

frontend.tsx

Agent completed work jubii

2026-07-09 08:07:32 +02:00

index.css

Run browser file transcription through the long-form pipeline when asked and the VAD model is available

2026-10-29 22:02:47 +01:00

index.html

Add typecheck + CI checks, light theme with self-hosted Inter, fix real-model test

2026-10-29 20:52:13 +01:00

index.ts

Add typecheck + CI checks, light theme with self-hosted Inter, fix real-model test

2026-10-29 20:52:13 +01:00