audiocraft-audio-generation

Generate music from text prompts using MusicGen by providing a natural-language description like “upbeat cinematic orchestral music with strings and brass” to produce a ready-to-review audio clip without manual composition.

Create sound effects from text for game audio and media production by using AudioGen to turn natural-language ideas into environmental audio and custom sound effect prototypes with controllable duration.

Yes, melody conditioning and stereo output are supported for text-to-music generation, alongside advanced controls like EnCodec compression, batch generation, temperature, and guidance scale handling.

Yes, you need PyTorch-compatible generation settings and torchaudio to run MusicGen and AudioGen, as the audio generation workflows require these frameworks to manage sample rates and audio processing.

Fine-tune MusicGen models for custom audio generation workflows by applying the deployment patterns and generation settings included in the Skill to adapt audio output style, quality, and duration to specific production needs.

EnCodec provides audio compression for production workflows within the AudioCraft ecosystem, but its application is limited to processing outputs generated through the MusicGen and AudioGen text-to-audio pipelines.