Home AI - Artificial Intelligence Stability AI Enhances Its Audio Generation Model for Optimal Performance on Arm Chips

Stability AI Enhances Its Audio Generation Model for Optimal Performance on Arm Chips

by admin

AI startup Stability AI has joined forces with chip manufacturer Arm to introduce Stability’s Stable Audio Open, an AI model capable of producing audio, including sound effects, specifically designed for mobile devices equipped with Arm processors.

While several AI-driven applications are able to create audio, such as Suno and Udio, many of them depend on cloud computing, which limits their offline usability. Additionally, certain audio generation models utilize copyrighted material for training, creating potential intellectual property concerns. Stability asserts that the training data for Stable Audio Open consists solely of royalty-free audio and music.

The Stable Audio Open model, optimized for Arm processors, will be showcased at the Mobile World Congress conference in Barcelona this week. This model is capable of generating sounds based on text prompts, such as, “Gentle ocean waves at sunset.” Stability claims to have collaborated with Arm to enhance and “distill” Stable Audio Open, achieving a 30x improvement in generation speed. An 11-second audio clip can be produced in approximately 8 seconds using an Armv9 CPU.

It is important to note that the optimized version of Stable Audio Open is currently not available for download. However, Stability’s CEO, Prem Akkaraju, indicated that the company aims to extend its models, including Stable Audio Open, to consumer applications and devices in the near future.

“As more professional creatives and organizations incorporate generative AI into their production processes, it’s crucial for our models and workflows to be widely accessible for builders and creators,” Akkaraju remarked. “We are thrilled to collaborate with Arm for this purpose.”

Stability states that it is working closely with Arm to further optimize and refine Stable Audio Open for mobile platforms.

Stability, known for the widely-used image generation model Stable Diffusion, secured additional funding last year as investors, including Eric Schmidt and Napster co-founder Sean Parker, sought to revitalize the company. The firm faced challenges under the leadership of co-founder and former CEO Emad Mostaque, whose management led to financial turmoil, staff departures, the collapse of a partnership with Canva, and growing concerns from investors regarding the company’s future.

Recently, Stability appointed a new CEO, brought on Titanic director James Cameron to its board, and launched several new image generation models.

Compiled by Techarena.au.
Fanpage: TechArena.au
Watch more about AI – Artificial Intelligence

You may also like

About Us

Get the latest tech news, reviews, and analysis on AI, crypto, security, startups, apps, fintech, gadgets, hardware, venture capital, and more.

Latest Articles