audiocraft-audio-generation - Generate music and sound with AudioCraft
Generates music from text, sound effects from text, and melody- or style-conditioned audio with AudioCraft models.
Tags
Updated: 2026-10-01Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Generate music from text
- Generate sound effects
- Generate melody-conditioned music
- Generate stereo audio
- Generate style-conditioned music
- Continue audio from input
- Compress audio with EnCodec
Inputs
- Text descriptions
- Melody audio
- Style reference audio
- Continuation audio
- Generation parameters
- Pretrained model checkpoints
Outputs
- Generated audio waveforms
- WAV audio files
- Stereo audio output
Requirements
- AudioCraft package
- PyTorch 2.0.0 or later
- Transformers 4.30.0 or later
- torchaudio