AudioCraft

Meta's audio generation library. MusicGen + AudioGen for music and sound effect generation.

Install

pip install audiocraft

Always review the official documentation before installing.

General-purpose speech recognition. Transcribes audio to text in 100+ languages.

Deep learning text-to-speech toolkit. 1100+ languages, multiple voices, fine-tunable models.

High-performance speech-to-text in C/C++. Runs Whisper models locally on CPU efficiently.

Text-prompted generative audio model. Generates speech, music, sound effects from text.