Activefrontieraudio Proprietary

Stable Audio 2.0

by Stability AI· Released April 2024

Stable Audio 2.0 is a generative audio model that produces high-quality, full-length music tracks (up to 3 minutes) from text prompts and optional reference audio. It introduces audio-to-audio generation, allowing users to transform existing audio clips, and supports stereo output at 44.1 kHz. The model is designed for musicians, content creators, and developers seeking efficient AI-powered music production.

Official Site API Docs

Input cost

—

Output cost

—

Context window

—

Max output

—

Modalities

audio

License

proprietary

Capabilities

Text-to-Audio GenerationAudio-to-Audio GenerationStereo Output44.1 kHz Sampling RateUp to 3-Minute Track LengthPrompt Conditioning

Best For

Generating full-length music tracks and transforming audio clips with text prompts.

Strengths

Generates up to 3 minutes of coherent music
Supports audio-to-audio transformation
High-quality stereo output at 44.1 kHz
Fast inference compared to prior models

Limitations

Limited to 3-minute outputs
May not perfectly replicate complex audio styles
Requires careful prompt engineering for best results
Not designed for speech or sound effects

Use Cases

Music production and composition

Background music for videos and games

Audio content creation for podcasts

Soundtrack generation for advertisements

Educational tools for music theory

Creative inspiration for musicians

Audio prototyping for media projects

Improvements Over Previous Model

Doubled maximum audio length from 90 seconds to 3 minutes
Introduced audio-to-audio generation capability
Improved audio quality and coherence
Added stereo output support
Enhanced prompt adherence and musical structure

Back to all models