Upload SpeechT5ForSpeechToText
Browse files- config.json +93 -0
- generation_config.json +9 -0
- model.safetensors +3 -0
config.json
ADDED
@@ -0,0 +1,93 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "microsoft/speecht5_asr",
|
3 |
+
"activation_dropout": 0.1,
|
4 |
+
"apply_spec_augment": true,
|
5 |
+
"architectures": [
|
6 |
+
"SpeechT5ForSpeechToText"
|
7 |
+
],
|
8 |
+
"attention_dropout": 0.1,
|
9 |
+
"bos_token_id": 0,
|
10 |
+
"conv_bias": false,
|
11 |
+
"conv_dim": [
|
12 |
+
512,
|
13 |
+
512,
|
14 |
+
512,
|
15 |
+
512,
|
16 |
+
512,
|
17 |
+
512,
|
18 |
+
512
|
19 |
+
],
|
20 |
+
"conv_kernel": [
|
21 |
+
10,
|
22 |
+
3,
|
23 |
+
3,
|
24 |
+
3,
|
25 |
+
3,
|
26 |
+
2,
|
27 |
+
2
|
28 |
+
],
|
29 |
+
"conv_stride": [
|
30 |
+
5,
|
31 |
+
2,
|
32 |
+
2,
|
33 |
+
2,
|
34 |
+
2,
|
35 |
+
2,
|
36 |
+
2
|
37 |
+
],
|
38 |
+
"decoder_attention_heads": 12,
|
39 |
+
"decoder_ffn_dim": 3072,
|
40 |
+
"decoder_layerdrop": 0.1,
|
41 |
+
"decoder_layers": 6,
|
42 |
+
"decoder_start_token_id": 2,
|
43 |
+
"encoder_attention_heads": 12,
|
44 |
+
"encoder_ffn_dim": 3072,
|
45 |
+
"encoder_layerdrop": 0.1,
|
46 |
+
"encoder_layers": 12,
|
47 |
+
"encoder_max_relative_position": 160,
|
48 |
+
"eos_token_id": 2,
|
49 |
+
"feat_extract_activation": "gelu",
|
50 |
+
"feat_extract_norm": "group",
|
51 |
+
"feat_proj_dropout": 0.0,
|
52 |
+
"guided_attention_loss_num_heads": 2,
|
53 |
+
"guided_attention_loss_scale": 10.0,
|
54 |
+
"guided_attention_loss_sigma": 0.4,
|
55 |
+
"hidden_act": "gelu",
|
56 |
+
"hidden_dropout": 0.1,
|
57 |
+
"hidden_size": 768,
|
58 |
+
"initializer_range": 0.02,
|
59 |
+
"is_encoder_decoder": true,
|
60 |
+
"layer_norm_eps": 1e-05,
|
61 |
+
"mask_feature_length": 10,
|
62 |
+
"mask_feature_min_masks": 0,
|
63 |
+
"mask_feature_prob": 0.0,
|
64 |
+
"mask_time_length": 10,
|
65 |
+
"mask_time_min_masks": 2,
|
66 |
+
"mask_time_prob": 0.05,
|
67 |
+
"max_length": null,
|
68 |
+
"max_speech_positions": 4000,
|
69 |
+
"max_text_positions": 600,
|
70 |
+
"model_type": "speecht5",
|
71 |
+
"num_conv_pos_embedding_groups": 16,
|
72 |
+
"num_conv_pos_embeddings": 128,
|
73 |
+
"num_feat_extract_layers": 7,
|
74 |
+
"num_mel_bins": 80,
|
75 |
+
"pad_token_id": 1,
|
76 |
+
"positional_dropout": 0.1,
|
77 |
+
"reduction_factor": 2,
|
78 |
+
"scale_embedding": false,
|
79 |
+
"speaker_embedding_dim": 512,
|
80 |
+
"speech_decoder_postnet_dropout": 0.5,
|
81 |
+
"speech_decoder_postnet_kernel": 5,
|
82 |
+
"speech_decoder_postnet_layers": 5,
|
83 |
+
"speech_decoder_postnet_units": 256,
|
84 |
+
"speech_decoder_prenet_dropout": 0.5,
|
85 |
+
"speech_decoder_prenet_layers": 2,
|
86 |
+
"speech_decoder_prenet_units": 256,
|
87 |
+
"torch_dtype": "float32",
|
88 |
+
"transformers_version": "4.45.1",
|
89 |
+
"unk_token_id": 3,
|
90 |
+
"use_cache": true,
|
91 |
+
"use_guided_attention_loss": false,
|
92 |
+
"vocab_size": 98
|
93 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 0,
|
4 |
+
"decoder_start_token_id": 2,
|
5 |
+
"eos_token_id": 2,
|
6 |
+
"max_length": 600,
|
7 |
+
"pad_token_id": 1,
|
8 |
+
"transformers_version": "4.45.1"
|
9 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:993b72d74757f59406dcff90fd58fa91e21b50af84493612000c7b923e102fe6
|
3 |
+
size 618919504
|