Meta has moved forward in the AI space rapidly. The Mark Zuckerberg-owned social media giant made its presence felt with the launch of its own "large open source language model" called LlaMa 2 to take on the likes of OpenAI, Google and Microsoft, and now taking things up a notch, Meta has unveiled its very own text-to-speech-based generative AI model called AudioCraft. Continue reading to know more about AudioCraft

Meta's AudioCraft generative AI model can help you generate high-quality music and audio using simple text-based prompts. AudioCraft's biggest USP is that it trains on RAW audio signals to deliver an authentic and realistic experience. This is similar to Google's audio AI tool, MusicLM.

AudioCraft is largely based on three distinct AI models: MusicGen, AudioGen and EnCodec. The purpose of MusicGen is to generate "music from text-based input", using Meta-owned and licensed music samples. AudioGen, on the other hand, generates "sound from text-based inputs" by using widely available sound effects. The EnCodec decoder is responsible for the generation of true-to-life audio outputs and, as Meta says, "with fewer artifacts."

This means you can easily generate different scenes with individually focused elements that will sync in the final output. For example, if you use the prompt "Jazz music from the 80's with a dog barking in the background", AudioCraft will use its MusicGen to deliver on your Jazz part while AudioGen will insert and mix the dog's barking in the background seamlessly. And all this will be presented to you via the advanced decoding functions of EnCodec. 🎵 Today we're sharing details about AudioCraft, a family of generative AI models that let you easily generate high-quality audio and music from text.https://t.co/04XAq4rlap pic.twitter.com/JreMIBGbTF— Meta Newsroom (@MetaNewsroom) August 2, 2023