mpv-whisper 0.2.0

Enhances a local MPV instance with Whisper transcription capabilities

MPV-Whisper

Adding Whisper audio transcription capabilities to MPV

Installing

Simply run the following to get started:

pip install mpv-whisper

To use the GPU, ensure that the appropriate version of PyTorch is installed.

Configuration

mpv-whisper searches for configuration in two places, and defaults back to the package-provided default if not found:

• ./mpv-whisper.toml
• ~/.config/mpv-whisper/config.toml
• package provided config.toml

Refer to the default config.toml for accepted configuration values.

Roadmap

• Using dump_cache or something similar to extract audio instead of running FFmpeg separately

Credits

• GhostNaN/whisper-subs for providing pointers on how to add whisper to MPV