jesse-c / local-audio-toolkit

Some handy tools to do with audio locally.

Audio Transcription and Summarization Tool

A command-line tool for transcribing and summarizing audio from various sources including YouTube videos, local MP3 files, and microphone recordings. Uses MLX Whisper for transcription and local LLM for summarization.

Usage

Important

Make sure you have LM Studio running locally on port 1234 for the summarization feature to work.

The tool provides these commands:

List Devices

uv run python src/main.py list-devices

Transcribe

Record from microphone:

uv run python src/main.py transcribe record

From YouTube:

uv run python src/main.py transcribe https://youtube.com/watch?v=...

From local MP3:

uv run python src/main.py transcribe path/to/audio.mp3

Summarise

Record from microphone:

uv run python src/main.py summarise record

From YouTube:

uv run python src/main.py summarise https://youtube.com/watch?v=...

From local MP3:

uv run python src/main.py summarise path/to/audio.mp3

With custom profile:

uv run python src/main.py summarise record --profile "software engineer writing PR description"

Building

1. Clone the repository
2. Install dependencies using uv:

uv sync --dev --all-extras

Testing

Tests can be run using pytest:

uv run pytest --cov=src tests/