Stability AI has launched Stable Audio 2.0, an update to its text-to-music generator, offering users extended capabilities in music creation. The new version allows users to generate tracks up to three minutes long at 44.1 KHz stereo, significantly enhancing their creative possibilities.
Audio-to-Audio Music Generation
One of the key highlights of Stable Audio 2.0 is its audio-to-audio generation feature, which empowers users to manipulate any audio sample using text-based AI prompts. This means that users can now transform their own audio samples into a wide array of sounds, opening up endless possibilities for experimentation and creativity.
However, it is important to note that Stability AI's terms of service require that any audio uploaded to the platform is free of copyrighted material. To ensure compliance, the tool employs a sophisticated content recognition filter.
The launch of Stable Audio 2.0 comes amidst some internal shakeup at Stability AI, notably the resignation of Ed Newton-Rex, the company's VP of audio.
Newton-Rex cited disagreements over the use of copyrighted works for AI training as the reason for his departure, emphasizing the importance of ethical considerations in AI development.
"To be clear, I'm a supporter of generative AI. It will have many benefits - that's why I've worked on it for 13 years. But I can only support generative AI that doesn't exploit creators by training models - which may replace them - on their work without permission," Newton-Rex said on X.
(Photo : Image via Stable AI)
Stability AI introduces Stable Audio 2.0, enhancing text-to-music capabilities, enabling creation of three-minute tracks at 44.1 KHz stereo, expanding creative options for users.
Stable Audio Trained on over 800,000 Audio Files
Unlike some other AI models in the market, Stable Audio and Stable Audio 2.0 are trained only on data licensed from the music library AudioSparx.
This library contains over 800,000 audio files, including music, sound effects, and single-instrument stems, along with text metadata. Musicians whose works are included in the library were given the option to opt out of being used to train Stable Audio's model.
Stable Audio 2.0 builds upon the success of its predecessor, Stable Audio 1.0, which was hailed as one of TIME's Best Inventions of 2023. The new version offers expanded functionalities, including full-length track generation with structured compositions, such as intros, progressions, and outros, as well as stereo sound effects.
More New Features
Users can now customize the final output to align with their specific needs, providing them with greater control and flexibility in their creative process.
This, coupled with the ability to generate sound and audio effects ranging from keyboard typing sounds to crowd roars, makes Stable Audio 2.0 a powerful tool for artists and musicians alike.
Stable Audio 2.0 is now available for free on the Stable Audio website, with plans for integration into the Stable Audio API in the near future.
Stay posted here at Tech Times.
Related Article : Beyoncé's 'Cowboy Carter' Album Takes Stand Against AI Music Production
