Stability AI has partnered with Arm to bring its Stable Audio Open model to mobile devices, enabling offline audio generation from text descriptions. Unlike other models that require cloud processing and may involve copyright issues, Stability claims their training set solely consists of royalty-free content. The optimized model, which is set to debut at Mobile World Congress, is reported to generate 11-second audio clips rapidly. CEO Prem Akkaraju highlights the importance of making generative AI tools accessible for creatives. Although the model isn't yet downloadable, plans for integration in consumer applications are forthcoming.
AI startup Stability AI has partnered with chipmaker Arm to optimize their Stable Audio Open model for mobile devices, allowing offline audio generation.
Stable Audio Open generates audio from text descriptions, claiming a training set composed entirely of royalty-free content, thus avoiding IP issues commonly associated with AI models.
Stability CEO Prem Akkaraju emphasized the need for generative AI tools like Stable Audio Open to be widely available for creatives and businesses.
The collaboration with Arm aims to accelerate audio generation speeds significantly while working towards a consumer-friendly release in the future.
Collection
[
|
...
]