hohsiangwu/wav2clip 🔢🖼️ → ❓

▶️ 896 runs 📅 Nov 2021 ⚙️ Cog 0.0.17 🔗 GitHub 📄 Paper ⚖️ License

audio-to-image image-generation vqgan-clip wav2clip

About

Image generation from Wav2CLIP through VQGAN-CLIP

Example Output

Output

[object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object][object Object]

Performance Metrics

288.30s Prediction Time

289.82s Total Time

All Input Parameters

{
  "seed": "-1",
  "audio": "https://replicate.delivery/mgxm/e0a846c0-2a54-46d0-bb04-8c71df2800b6/1-13572-A-46.wav",
  "iterations": "300",
  "learning_rate": "0.1001"
}

Input Parameters

seed Type: integerDefault: -1: Random seed, -1 for random
audio Type: stringDefault: null: Input audio file
iterations Type: integerDefault: 200Range: 1 - 300: Number of iterations
learning_rate Type: numberDefault: 0.1Range: 0.0001 - 0.5: Learning rate

Output Schema

Type: array • Items Type: object

Example Execution Logs

Using seed: 42626
Iteration 0; loss: 1.18344
Iteration 10; loss: 1.10417
Iteration 20; loss: 1.08918
Iteration 30; loss: 1.02602
Iteration 40; loss: 0.964155
Iteration 50; loss: 0.929151
Iteration 60; loss: 0.948758
Iteration 70; loss: 1.00458
Iteration 80; loss: 0.99745
Iteration 90; loss: 1.03451
Iteration 100; loss: 0.945668
Iteration 110; loss: 0.853544
Iteration 120; loss: 0.981887
Iteration 130; loss: 0.84142
Iteration 140; loss: 0.852543
Iteration 150; loss: 0.864299
Iteration 160; loss: 0.915457
Iteration 170; loss: 0.942693
Iteration 180; loss: 0.805264
Iteration 190; loss: 0.814359
Iteration 200; loss: 0.945025
Iteration 210; loss: 0.90828
Iteration 220; loss: 0.784327
Iteration 230; loss: 0.90776
Iteration 240; loss: 0.853381
Iteration 250; loss: 0.866105
Iteration 260; loss: 0.779212
Iteration 270; loss: 0.803184
Iteration 280; loss: 0.924357
Iteration 290; loss: 0.86961

Version Details

Version ID: c28f117f407cedf7dad1b6eb6dec3f2e7a11a74e0470935c81450c57116cc6fb
Version Created: November 29, 2021

Run on Replicate →