transcribe-voice-cli

Simple CLI tool to transcribe audio files into text files.

It tries:

1. faster-whisper (CPU int8)
2. falls back to openai-whisper

Features

• Input audio path as argument
• Output transcript path configurable (-o/--output)
• Optional language override (--language)
• Engine selection (--engine auto|faster-whisper|openai-whisper)
• Segment timestamps included in output

Install

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Usage

python transcribe_voice.py /path/to/audio.m4a

Custom output:

python transcribe_voice.py /path/to/audio.m4a -o /path/to/transcript.txt

Force engine/language:

python transcribe_voice.py /path/to/audio.m4a --engine faster-whisper --language de

Notes

• Default output file is transcript_<input_stem>.txt in the same folder as input.
• faster-whisper uses CPU int8 by default.