AudioCraft is a Meta AI research framework for generating music and sound effects from text, and for audio compression.
AudioCraft is a unified codebase developed by Meta AI for generative audio tasks, including music, sound effects, and audio compression. It simplifies the design of generative audio models by using a single autoregressive Language Model (LM) that processes streams of compressed discrete audio representations, or tokens. The framework leverages the EnCodec neural audio codec to convert raw audio signals into discrete tokens, which are then modeled by the LM. The generated tokens are subsequently decoded back into audio waveforms by EnCodec. AudioCraft encompasses two primary models: MusicGen, which generates diverse music samples from text inputs, and AudioGen, which focuses on text-to-sound generation for environmental sounds. The system allows for control over generation through conditioning models, such as pretrained text encoders for text-to-audio applications. It aims to efficiently model audio sequences, capturing long-term dependencies and producing high-quality audio.
Platforms: Web
View full AudioCraft profile on Tools-Radar | Browse Voice & Audio tools | Alternatives to AudioCraft
Tools-Radar is a free directory of 10,000+ AI tools — discover, compare, and choose the right AI software for your needs. Visit tools-radar.com