May 15, 2011 - TechCrunch released a blog post yesterday (May 14), reporting that AI startups Stability AI Launches Stable Audio Open Small.Claims to be the fastest "stereo" audio generation AI model on the market, running on smartphones.
Compared to other AI audio apps like Suno and Udio, Stable Audio Open Small doesn't need to rely on cloud-based processing and its efficient design allows it to run smoothly on smartphones, says Stability AI.The model generates audio clips of up to 11 seconds on your phone in less than 8 seconds, making it ideal for quickly creating sound effects such as drum beats or instrument clips.
In terms of training, Stability AI has partnered with chip maker Arm and is optimized for Arm CPUs, which can be run offline on smartphones; in terms of the training dataset, unlike competitors such as Suno and Udio (which allegedly use data containing copyrighted content, posing an intellectual property risk), it's all sourced from the Free Music Archive and Freesound royalty-free audio libraries. Freesound royalty-free audio libraries.
Note: At this stage Stable Audio Open Small English tips only, cannot generate realistic vocals or high-quality songs, and because the training data is biased toward Western styles, theMixed performance across music genres.
The terms of use are also strict: researchers, hobbyists, and businesses with less than $1 million in annual revenue can use it for free, but developers or organizations above that revenue threshold will need to purchase an enterprise license for Stability AI.