AudioCraft, Meta’s open source AI tool for creating music/audio, allows users to achieve this through text prompts

According to news on August 2, Eastern Time, Meta Company open sourced a generative AI tool called AudioCraft on August 2, which can help users create music and audio through text prompts

According to Meta’s official introduction, AudioCraft contains three core components:

• MusicGen: Trained using Meta owned/specially licensed music to generate music based on text prompts.
• AudioGen: Use public sound effects for training to generate audio or expand existing audio. Later, you can also generate environmental sound effects (such as dog barking, car horns, footsteps on the wooden floor).
• EnCodec (improved version): Neural network-based audio compression codec that can generate higher quality music and reduce artifacts, or perform lossless compression of audio files.

After the tool is open sourced, relevant researchers and practitioners can use their own data sets to train models. Officials claim that the AudioCraft series models can produce high-quality audio stably over the long term and are easy to use, can "provide inspiration" to musicians and sound designers, helping them quickly brainstorm ideas and "iterate their work in new ways." .

According to previous reports from IT House, In June this year, Meta open sourced the AI ​​language model MusicGen, which is based on the Transformer model launched by Google in 2017. As the name of the model suggests, MusicGen is mainly used for music generation. It can convert text and existing melodies into complete music pieces.

Video demonstration:

AudioCraft project address: Click here to go

The above is the detailed content of AudioCraft, Meta’s open source AI tool for creating music/audio, allows users to achieve this through text prompts. For more information, please follow other related articles on the PHP Chinese website!