WhisperSubs is a Python tool that transcribes video/audio files to subtitles. It uses the faster-whisper library for transcription (based on OpenAI's Whisper ASR model), and PyAV (ffmpeg bindings) for audio/video demuxing and decoding.
Key features included in this initial release:
- transcription of audio or video files.
- automatic language detection by whisper model
- translation of the detected language to English.
- GPU acceleration support with CUDA.
- command-line interface.
- easy installation via pipx.
Refer to the README for installation and usage instructions.