Appgithub-com-pierrunoyt-cohere-transcribe-pinokio

Download Pinokio

Sign in Create accountJoin

Sign in Create accountJoin

Installable

Cohere Transcribe

github.com/PierrunoYT/cohere-transcribe-pinokioUpdated 17d ago

State-of-the-art open-source speech recognition model supporting 14 languages. 2B parameter ASR model from Cohere Labs.

Loading repository owner…

Loading community details…

Loading project activity…

Community

Sort byCreate post

Ask or share about Cohere Transcribe…

Loading...

Pinokio

Check-ins (1)

1 check-in

Platforms (1)

Windows100.0%

Arch (1)

x64100.0%

GPU (1)

nvidia geforce rtx 4090100.0%

RAM (1)

47GB100.0%

VRAM (1)

24GB100.0%

Recent commits

fix: keep a running server reachable in the Pinokio menu

PierrunoYT2 weeks ago4f77689

fix: decode m4a, aac, webm and other containers via ffmpeg

PierrunoYT2 weeks ago333f8dd

fix: report transcription failures instead of a bare "Error"

PierrunoYT2 weeks agob399ad8

fix: reject near-empty audio and skip sliver chunks

PierrunoYT2 weeks ago4217296

fix: validate installs and refresh platform dependencies on update

PierrunoYTlast monthbfae9ec

fix: bound long-form inference to one audio chunk at a time

PierrunoYTlast month7885c37

fix: stabilize model caching and transcription callbacks

PierrunoYTlast monthf20b36a

Fix Pinokio install/torch bugs found in script audit

PierrunoYT4 months ago943ad14

feat: add bundle requirement to install.js and remove local server name from start.js

Pierre Bruno5 months ago7d860ff

refactor: remove redundant title/description/icon from pinokio.js (head lives in pinokio.json)

PierrunoYT5 months ago6956a21