declare-lab/tangoflux 🔢📝 → 🖼️

▶️ 51.3K runs 📅 Dec 2024 ⚙️ Cog 0.9.23 🔗 GitHub 📄 Paper ⚖️ License
sound-effect-generation text-to-audio

About

Super Fast and Faithful Text to Audio Generation with Flow Matching and Clap-Ranked Preference Optimization

Example Output

Prompt:

"The deep growl of an alligator ripples through the swamp as reeds sway with a soft rustle and a turtle splashes into the murky water"

Output

Example output

Performance Metrics

1.39s Prediction Time
1.39s Total Time
All Input Parameters
{
  "steps": 25,
  "prompt": "The deep growl of an alligator ripples through the swamp as reeds sway with a soft rustle and a turtle splashes into the murky water",
  "duration": 10,
  "guidance_scale": 4.5
}
Input Parameters
steps Type: integerDefault: 25Range: 1 - 200
Number of inference steps
prompt Type: stringDefault: Hammer slowly hitting the wooden table
Input prompt
duration Type: integerDefault: 10
Duration of the output audio in seconds
guidance_scale Type: numberDefault: 4.5Range: 1 - 20
Scale for classifier-free guidance
Output Schema

Output

Type: stringFormat: uri

Example Execution Logs
0%|          | 0/25 [00:00<?, ?it/s]
0%|          | 0/25 [00:01<?, ?it/s]
Version Details
Version ID
fcdc421786888a045329d7c4e1874764433a2516b21f4c34bd3da4e054d04cf9
Version Created
December 31, 2024
Run on Replicate →