Official

google / lyria-2

Lyria 2 is a music generation model that produces 48kHz stereo audio through text-based prompts

  • Public
  • 5.2K runs
  • $2 per thousand seconds of output audio
  • Commercial use
Iterate in playground

Input

*string
Shift + Return to add a new line

Text prompt for audio generation

string
Shift + Return to add a new line

Description of what to exclude from the generated audio

integer

Random seed. Omit for random generations

Output

Video Player is loading.
Current Time 00:00:000
Duration 00:00:000
Loaded: 0%
Stream Type LIVE
Remaining Time 00:00:000
 
1x
Generated in

Pricing

Model pricing for google/lyria-2. Looking for volume pricing? Get in touch.

$2
per thousand seconds of output audio

or around 500 seconds for $1

Official models are always on, maintained, and have predictable pricing. Learn more.

Check out our docs for more information about how pricing works on Replicate.

Readme

Lyria 2

Lyria 2 is an AI model for music generation that produces high-fidelity audio output at 48kHz in stereo format.

Technical Specifications

  • 30 seconds of 48kHz stereo audio
  • Professional-grade audio suitable for direct project integration
  • Implements SynthID for audio watermarking

Genre Coverage

  • Classical
  • Jazz
  • Pop
  • Electronic
  • Orchestral
  • Various regional and fusion styles

Limitations

Current known limitations include:

  • Restricted language support
  • Limited genre coverage in certain musical styles
  • Model performance varies across different musical complexities