google/lyria-2 🔢📝 → 🖼️

⭐ Official ▶️ 54.4K runs 📅 May 2025 ⚙️ Cog 0.16.9
audio-watermarking music-generation text-to-audio text-to-music

About

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

Example Output

Prompt:

"Futuristic country music, steel guitar, huge 808s, synth wave elements space western cosmic twang soaring vocals"

Output

Example output

Performance Metrics

26.17s Prediction Time
34.54s Total Time
Input Parameters
seed Type: integer
Random seed. Omit for random generations
prompt (required) Type: string
Text prompt for audio generation
negative_prompt Type: string
Description of what to exclude from the generated audio
Output Schema

Output

Type: string • Format: uri

Example Execution Logs 
Using seed: 411153273
Generating audio...
Generated audio in 26.08 seconds
Version Details
Version ID
bb621623ee2772c96d300b2a303c9e444b482f6b0fafcc7424923e1429971120
Version Created
November 25, 2025
Run on Replicate →