Meta AudioCraft

Voice & Multimodal
★ 23.3k stars

Meta AudioCraft is an open-source suite for generating audio, featuring MusicGen and AudioGen models that accept detailed text prompts for fine-grained control over output.

Visit on GitHub ↗ facebookresearch/audiocraft
GitHub repository github.com/facebookresearch/audiocraft
Stars 23,322
Category Voice & Multimodal

More in Voice & Multimodal

Whisper v3 (local) ★ 100.8k

Whisper v3 runs locally to transcribe speech to text, supporting 99 languages with a large…
Bark ★ 39.1k

Bark is an open-source text-to-audio model that generates speech, music, and sound effects…
Rasa ★ 21.2k

Rasa is an open-source conversational AI framework that enables developers to train custom…
Pipecat ★ 12.5k

Pipecat is a production-grade voice AI framework that delivers sub-250ms latency, supports…
LiveKit Agents ★ 10.7k

LiveKit Agents is a framework for building real-time, multimodal AI agents that communicat…
Kokoro ★ 7.2k

Kokoro is an open-weight text-to-speech model with 82 million parameters, optimized for fa…