Facebook has released its AI audio tools, called AudioCraft, designed for generative AI for audio and music.

Imagine being able to experiment with new compositions as a professional musician without having to play a single note on an instrument. Or a small business owner easily adding music to their most recent Instagram video advertisement. Our most recent AI tool, AudioCraft, which creates high-quality, realistic audio and music from text, makes that promise.

MusicGen, AudioGen, and EnCodec are the three models that make up AudioCraft. While AudioGen, which was trained on public sound effects, generates audio from text prompts, MusicGen, which was trained with Meta-owned and expressly licenced music, produces music from text prompts.

In Meta statement.,“We're pleased to announce the availability of an updated version of our EnCodec decoder, which enables the creation of music of greater quality and with fewer artefacts. Our pre-trained AudioGen models are now available, allowing you to create environmental noises and sound effects like a dog barking, a car honking, or footfall on a wooden floor. The entire code and weights for the AudioCraft model are also included.”

It was stated in the article that "We're open-sourcing these models, giving researchers and practitioners access so they can train their own models with their own datasets for the first time, and help advance the field of AI-generated audio and music."

said Meta While generative AI for photos, video, and text has generated a lot of buzz, audio has appeared to lag behind. There is some work available, but it is quite complex and not particularly open, so people are unable to experiment with it easily. Any type of high-fidelity audio production demands the modelling of intricate signals and patterns at various scales. Since it consists of both local and long-range patterns, ranging from a set of notes to a global musical structure with several instruments, music is possibly the most difficult sort of audio to produce.

The devices in the AudioCraft family are simple to operate and can produce audio of a high standard over an extended period of time. In comparison to earlier research in the field, we have simplified the design of generative models for audio with AudioCraft. We provide users with the full recipe to experiment with the models that Meta has been building over the past few years, as well as the freedom to push the boundaries and create their own models.

Music, sound, compression, and creation are all handled by AudioCraft in one location. People that want to create better sound generators, compression algorithms, or music generators can do so in the same code base and expand on previous work because it is simple to reuse and build upon.

A strong open source foundation will encourage innovation and enhance how we create and consume music and audio in the future. We believe that MusicGen has the potential to develop into a new class of instrument with more controls, similar to how synthesisers did when they originally arrived.

The AudioCraft family of models, according to Meta, "we see as tools for musicians and sound designers to bring inspiration, let people quickly brainstorm and iterate on their works in innovative ways. We eagerly anticipate what users will produce using Audiocraft.

Learn more about AudioCraft on their AI blog.

Please send press releases to [email protected].Other ways to contact us. Editorially, we may rewrite headlines and descriptions.

Recommend Podworks.io: Hire the Perfect Freelancer for your Podcast.

Podworks.io is home to the world's most comprehensive Podcast job board and candidate board.

Want your positions to be featured on our Podworks.io Jobs Board?