cat ~/projects/cohere-transcribe.md
cohere-transcribe
High-throughput Arabic/English transcription, batched properly.
A Python package for offline batch transcription with Cohere's 2B Arabic/English ASR model: optimised voice activity detection, multi-file GPU batching under bounded memory, subtitle output and optional word-level timestamps.
- Role
- Contributor
- Year
- 2026
- Status
- Live
- type
- AI
- Built with
- PythonSilero VADGPU batchingSRT / VTTHugging Face
- Where to find it
- Source
head -n 40 overview
What it is
Transcribing one file is a solved problem. Transcribing a directory of hundreds without running out of GPU memory, and without spending most of the run on silence, is not.
The package handles individual files, multiple paths and nested directories with bounded-memory batching, and can return results directly or publish plain text, approximate segment-timed subtitles, or word-timed subtitles.
diff before after
Challenges
A line taken out, a line put in. The problem first, then what actually replaced it.
Most of a recording is silence
Feeding raw audio to the model spends a large share of the compute transcribing pauses, which is the main reason long-file throughput is bad.
Voice activity detection segments speech first, so only speech reaches the model. The batching is then built around those segments rather than around file boundaries.
tail outcomes
Results
- Directory-scale batch transcription with bounded memory and subtitle output.
Keep readingتابع القراءة
opent3-code
Next projectT3 CodeControl the coding agents on your machine, from anywhere.