Stability AI has released Stable Audio 2.0, adding key new features to the company's text-to-music generator. Now, users can create tracks of up to three minutes in 44.1KHz stereo from a natural language prompt like “A beautiful piano arpeggio grows into a full beautiful orchestral piece” or “Lo-fi funk.”
Stable Audio 2.0 also features an audio-to-audio generation feature, allowing users to manipulate any audio sample they want using text-based AI prompts. However, its terms of use require that any audio uploaded to this tool be free of copyrighted material, with the tool using a content recognition filter to ensure compliance.
Stable Audio 2.0 is the company's first major music product launched by audio vp, Ed Newton-Rex, resigned. To announce his departure, Newton-Rex wrote a lengthy post on social media saying“I do not agree with the company's view that training AI production models on copyrighted works is 'fair use'… I hope others will speak up, either internally or publicly, so that companies realize that exploiting creators is not may be a long-term solution to genetic artificial intelligence.”
Unlike some of the company's other models, Stable Audio and Stable Audio 2.0 are only trained on training data licensed from the AudioSparx music library. The library contains over 800,000 audio files containing music, sound effects and stems of an instrument as well as text metadata. All musicians who created the works in the AudioSparx library had the option to opt out of using them to train the Stable Audio model.
Stable Audio 2.0 is now available for free on Stable Audio Website and will soon be available in the Stable Audio API.
from our partners at https://www.billboard.com/business/tech/stability-ai-text-to-music-generator-stable-audio-update-1235647936/
