Upload Wav2Vec2ForCTC
Browse files- README.md +3 -3
- config.json +7 -8
- model.safetensors +2 -2
README.md
CHANGED
@@ -1,13 +1,13 @@
|
|
1 |
---
|
|
|
|
|
|
|
2 |
language:
|
3 |
- eng
|
4 |
license: apache-2.0
|
5 |
-
base_model: facebook/wav2vec2-base-960h
|
6 |
tags:
|
7 |
- '[finetuned_model, lj_speech11]'
|
8 |
- generated_from_trainer
|
9 |
-
datasets:
|
10 |
-
- FYP/LJ-Speech1111LJ
|
11 |
model-index:
|
12 |
- name: SpeechT5 STT Wav2Vec2
|
13 |
results: []
|
|
|
1 |
---
|
2 |
+
base_model: facebook/wav2vec2-base-960h
|
3 |
+
datasets:
|
4 |
+
- FYP/LJ-Speech1111LJ
|
5 |
language:
|
6 |
- eng
|
7 |
license: apache-2.0
|
|
|
8 |
tags:
|
9 |
- '[finetuned_model, lj_speech11]'
|
10 |
- generated_from_trainer
|
|
|
|
|
11 |
model-index:
|
12 |
- name: SpeechT5 STT Wav2Vec2
|
13 |
results: []
|
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "facebook/wav2vec2-
|
3 |
"activation_dropout": 0.1,
|
4 |
"adapter_attn_dim": null,
|
5 |
"adapter_kernel_size": 3,
|
@@ -50,16 +50,15 @@
|
|
50 |
"feat_extract_activation": "gelu",
|
51 |
"feat_extract_dropout": 0.0,
|
52 |
"feat_extract_norm": "group",
|
53 |
-
"feat_proj_dropout": 0.
|
54 |
"feat_quantizer_dropout": 0.0,
|
55 |
"final_dropout": 0.1,
|
56 |
-
"gradient_checkpointing": false,
|
57 |
"hidden_act": "gelu",
|
58 |
"hidden_dropout": 0.1,
|
59 |
"hidden_dropout_prob": 0.1,
|
60 |
-
"hidden_size":
|
61 |
"initializer_range": 0.02,
|
62 |
-
"intermediate_size":
|
63 |
"layer_norm_eps": 1e-05,
|
64 |
"layerdrop": 0.1,
|
65 |
"mask_feature_length": 10,
|
@@ -70,15 +69,15 @@
|
|
70 |
"mask_time_prob": 0.05,
|
71 |
"model_type": "wav2vec2",
|
72 |
"num_adapter_layers": 3,
|
73 |
-
"num_attention_heads":
|
74 |
"num_codevector_groups": 2,
|
75 |
"num_codevectors_per_group": 320,
|
76 |
"num_conv_pos_embedding_groups": 16,
|
77 |
"num_conv_pos_embeddings": 128,
|
78 |
"num_feat_extract_layers": 7,
|
79 |
-
"num_hidden_layers":
|
80 |
"num_negatives": 100,
|
81 |
-
"output_hidden_size":
|
82 |
"pad_token_id": 0,
|
83 |
"proj_codevector_dim": 256,
|
84 |
"tdnn_dilation": [
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "facebook/wav2vec2-large-960h",
|
3 |
"activation_dropout": 0.1,
|
4 |
"adapter_attn_dim": null,
|
5 |
"adapter_kernel_size": 3,
|
|
|
50 |
"feat_extract_activation": "gelu",
|
51 |
"feat_extract_dropout": 0.0,
|
52 |
"feat_extract_norm": "group",
|
53 |
+
"feat_proj_dropout": 0.0,
|
54 |
"feat_quantizer_dropout": 0.0,
|
55 |
"final_dropout": 0.1,
|
|
|
56 |
"hidden_act": "gelu",
|
57 |
"hidden_dropout": 0.1,
|
58 |
"hidden_dropout_prob": 0.1,
|
59 |
+
"hidden_size": 1024,
|
60 |
"initializer_range": 0.02,
|
61 |
+
"intermediate_size": 4096,
|
62 |
"layer_norm_eps": 1e-05,
|
63 |
"layerdrop": 0.1,
|
64 |
"mask_feature_length": 10,
|
|
|
69 |
"mask_time_prob": 0.05,
|
70 |
"model_type": "wav2vec2",
|
71 |
"num_adapter_layers": 3,
|
72 |
+
"num_attention_heads": 16,
|
73 |
"num_codevector_groups": 2,
|
74 |
"num_codevectors_per_group": 320,
|
75 |
"num_conv_pos_embedding_groups": 16,
|
76 |
"num_conv_pos_embeddings": 128,
|
77 |
"num_feat_extract_layers": 7,
|
78 |
+
"num_hidden_layers": 24,
|
79 |
"num_negatives": 100,
|
80 |
+
"output_hidden_size": 1024,
|
81 |
"pad_token_id": 0,
|
82 |
"proj_codevector_dim": 256,
|
83 |
"tdnn_dilation": [
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:29efe5b81b4080d96694ddee2154e817419674ce8286162e8b712130d20cf306
|
3 |
+
size 1261897368
|