Stability AI Launches Stable Audio: A Text-to-Audio Generative AI Platform

Stability AI, known for AI-generated visuals, introduces Stable Audio – a text-to-audio generative AI platform. It uses diffusion models trained with audio to create songs and background audio for various projects.

Unlike traditional audio diffusion models, Stable Audio offers flexibility in audio length, improving usability for music production. Users can control song duration with more precision.

Stability AI trained Stable Audio on a vast dataset of over 800,000 audio files, including music, sound effects, and single-instrument stems, alongside text metadata from AudioSparx. This extensive dataset ensures a rich audio generation experience.

Stable Audio offers three pricing tiers: Free (up to 45 seconds of audio for 20 tracks/month), Professional ($11.99/month for 500 tracks up to 90 seconds), and Enterprise (customizable for businesses). Free users cannot use generated audio commercially.

Text-to-audio generation is gaining traction in generative AI, with Meta and Google also exploring similar concepts. Such platforms streamline content creation, particularly for podcasts and videos.

Stability AI continues its expansion into audio generation, video, and 3D images, broadening the creative possibilities with AI-driven content.