Delete model-index.json
Browse files- model-index.json +0 -68
model-index.json
DELETED
|
@@ -1,68 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"name": "MaskGCT",
|
| 3 |
-
"_name_or_path": "amphion/MaskGCT",
|
| 4 |
-
"modelId": "amphion/MaskGCT",
|
| 5 |
-
"architectures": [
|
| 6 |
-
"MaskGCTModel"
|
| 7 |
-
],
|
| 8 |
-
"model_type": "maskgct",
|
| 9 |
-
"task_specific_params": {
|
| 10 |
-
"text-to-speech": {
|
| 11 |
-
"supported_tasks": [
|
| 12 |
-
"zero-shot-tts",
|
| 13 |
-
"non-autoregressive-tts"
|
| 14 |
-
]
|
| 15 |
-
}
|
| 16 |
-
},
|
| 17 |
-
"tags": [
|
| 18 |
-
"text-to-speech",
|
| 19 |
-
"safetensors"
|
| 20 |
-
],
|
| 21 |
-
"pipeline_tag": "text-to-speech",
|
| 22 |
-
"language": [
|
| 23 |
-
"en",
|
| 24 |
-
"zh"
|
| 25 |
-
],
|
| 26 |
-
"license": "cc-by-nc-4.0",
|
| 27 |
-
"datasets": [
|
| 28 |
-
"Emilia-100k"
|
| 29 |
-
],
|
| 30 |
-
"model_structure": {
|
| 31 |
-
"semantic_codec": {
|
| 32 |
-
"type": "w2v-bert-2.0",
|
| 33 |
-
"description": "Converting speech to semantic tokens"
|
| 34 |
-
},
|
| 35 |
-
"acoustic_codec": {
|
| 36 |
-
"type": "codec",
|
| 37 |
-
"description": "Converting speech to acoustic tokens and reconstructing waveform"
|
| 38 |
-
},
|
| 39 |
-
"maskgct_t2s": {
|
| 40 |
-
"type": "transformer",
|
| 41 |
-
"description": "Predicting semantic tokens with text and prompt semantic tokens"
|
| 42 |
-
},
|
| 43 |
-
"maskgct_s2a": {
|
| 44 |
-
"variants": {
|
| 45 |
-
"1layer": {
|
| 46 |
-
"type": "transformer",
|
| 47 |
-
"description": "Single layer model for acoustic token prediction"
|
| 48 |
-
},
|
| 49 |
-
"full": {
|
| 50 |
-
"type": "transformer",
|
| 51 |
-
"description": "Full model for acoustic token prediction"
|
| 52 |
-
}
|
| 53 |
-
},
|
| 54 |
-
"description": "Predicts acoustic tokens conditioned on semantic tokens"
|
| 55 |
-
}
|
| 56 |
-
},
|
| 57 |
-
"training_data": {
|
| 58 |
-
"total_hours": 100000,
|
| 59 |
-
"language_split": {
|
| 60 |
-
"english": 50000,
|
| 61 |
-
"chinese": 50000
|
| 62 |
-
}
|
| 63 |
-
},
|
| 64 |
-
"paper": {
|
| 65 |
-
"title": "MaskGCT: Zero-Shot Text-to-Speech with Masked Generative Codec Transformer",
|
| 66 |
-
"arxiv_id": "2409.00750"
|
| 67 |
-
}
|
| 68 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|