cjwbw/melotts 📝🔢❓ → 🖼️

▶️ 1.4K runs 📅 Mar 2024 ⚙️ Cog 0.9.4 🔗 GitHub ⚖️ License

multilingual text-to-speech

About

High-quality multilingual text-to-speech library

Example Output

Output

Performance Metrics

2.88s Prediction Time

2.90s Total Time

All Input Parameters

{
  "text": "The field of text-to-speech has seen rapid development recently.",
  "speed": 1,
  "speaker": "EN-BR",
  "language": "EN"
}

Input Parameters

text Type: stringDefault: The field of text-to-speech has seen rapid development recently.
speed Type: numberDefault: 1Range: 0.1 - 10: Speed of the output.
speaker Default: EN-US: For EN, choose a speaker, for other langauges, leave it blank.
language Default: EN

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

> Text split to sentences.
The field of text-to-speech has seen rapid development recently.
> ===========================
  0%|          | 0/1 [00:00<?, ?it/s]
model.safetensors:   0%|          | 0.00/440M [00:00<?, ?B/s][A
model.safetensors:   0%|          | 47.8k/440M [00:00<19:35, 375kB/s][A
model.safetensors:   2%|▏         | 10.5M/440M [00:00<00:13, 32.6MB/s][A
model.safetensors:   7%|▋         | 31.5M/440M [00:00<00:05, 81.1MB/s][A
model.safetensors:  19%|█▉        | 83.9M/440M [00:00<00:01, 199MB/s] [A
model.safetensors:  33%|███▎      | 147M/440M [00:00<00:00, 312MB/s] [A
model.safetensors:  69%|██████▉   | 304M/440M [00:00<00:00, 659MB/s][A
model.safetensors: 100%|█████████▉| 440M/440M [00:00<00:00, 462MB/s]
Some weights of the model checkpoint at bert-base-uncased were not used when initializing BertForMaskedLM: ['bert.pooler.dense.bias', 'bert.pooler.dense.weight', 'cls.seq_relationship.bias', 'cls.seq_relationship.weight']
- This IS expected if you are initializing BertForMaskedLM from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
- This IS NOT expected if you are initializing BertForMaskedLM from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).
100%|██████████| 1/1 [00:02<00:00,  2.49s/it]
100%|██████████| 1/1 [00:02<00:00,  2.49s/it]

Version Details

Version ID: 2e4d356f3715d98c183ef097ce2cf410def83ca9fbbdd5f8a32ba056123e6a6f
Version Created: March 3, 2024

Run on Replicate →