Upload model (#8)
Browse files- Upload model (19ebaa199dd23d3a87c50163396bdcef0eeef474)
- config.json +5 -34
config.json
CHANGED
@@ -3,7 +3,6 @@
|
|
3 |
"activation_function": "relu",
|
4 |
"adaptor_dropout": 0.1,
|
5 |
"adaptor_kernel_size": 8,
|
6 |
-
"adaptor_layer_norm": true,
|
7 |
"adaptor_stride": 8,
|
8 |
"add_adapter": true,
|
9 |
"architectures": [
|
@@ -11,35 +10,8 @@
|
|
11 |
],
|
12 |
"attention_dropout": 0.1,
|
13 |
"bos_token_id": 2,
|
14 |
-
"
|
15 |
"conv_depthwise_kernel_size": 31,
|
16 |
-
"conv_dim": [
|
17 |
-
512,
|
18 |
-
512,
|
19 |
-
512,
|
20 |
-
512,
|
21 |
-
512,
|
22 |
-
512,
|
23 |
-
160
|
24 |
-
],
|
25 |
-
"conv_kernel": [
|
26 |
-
10,
|
27 |
-
3,
|
28 |
-
3,
|
29 |
-
3,
|
30 |
-
3,
|
31 |
-
2,
|
32 |
-
2
|
33 |
-
],
|
34 |
-
"conv_stride": [
|
35 |
-
5,
|
36 |
-
2,
|
37 |
-
2,
|
38 |
-
2,
|
39 |
-
2,
|
40 |
-
2,
|
41 |
-
2
|
42 |
-
],
|
43 |
"decoder_attention_heads": 16,
|
44 |
"decoder_ffn_dim": 4096,
|
45 |
"decoder_layerdrop": 0.05,
|
@@ -51,26 +23,22 @@
|
|
51 |
"encoder_layerdrop": 0.05,
|
52 |
"encoder_layers": 12,
|
53 |
"eos_token_id": 3,
|
54 |
-
"
|
55 |
"hidden_size": 1024,
|
56 |
-
"init_std": 0.02,
|
57 |
"initializer_range": 0.02,
|
58 |
"is_encoder_decoder": true,
|
59 |
"lang_embed_dim": 256,
|
60 |
"layer_norm_eps": 1e-05,
|
61 |
-
"layerdrop": 0.1,
|
62 |
"leaky_relu_slope": 0.1,
|
63 |
"max_new_tokens": 256,
|
64 |
"max_position_embeddings": 4096,
|
65 |
"max_source_positions": 4096,
|
66 |
-
"model_in_dim": 1792,
|
67 |
"model_type": "seamless_m4t",
|
68 |
"num_adapter_layers": 1,
|
69 |
"num_attention_heads": 16,
|
70 |
"num_conv_pos_embedding_groups": 16,
|
71 |
"num_conv_pos_embeddings": 128,
|
72 |
"num_hidden_layers": 12,
|
73 |
-
"output_hidden_size": null,
|
74 |
"pad_token_id": 0,
|
75 |
"position_embeddings_type": "relative",
|
76 |
"resblock_dilation_sizes": [
|
@@ -102,6 +70,7 @@
|
|
102 |
"speech_encoder_dropout": 0.0,
|
103 |
"speech_encoder_hidden_act": "swish",
|
104 |
"speech_encoder_intermediate_size": 4096,
|
|
|
105 |
"speech_encoder_layers": 12,
|
106 |
"spkr_embed_dim": 256,
|
107 |
"t2u_bos_token_id": 0,
|
@@ -114,7 +83,9 @@
|
|
114 |
"t2u_encoder_layers": 4,
|
115 |
"t2u_eos_token_id": 2,
|
116 |
"t2u_max_new_tokens": 1024,
|
|
|
117 |
"t2u_num_langs": 38,
|
|
|
118 |
"t2u_pad_token_id": 1,
|
119 |
"torch_dtype": "float32",
|
120 |
"transformers_version": "4.33.0.dev0",
|
|
|
3 |
"activation_function": "relu",
|
4 |
"adaptor_dropout": 0.1,
|
5 |
"adaptor_kernel_size": 8,
|
|
|
6 |
"adaptor_stride": 8,
|
7 |
"add_adapter": true,
|
8 |
"architectures": [
|
|
|
10 |
],
|
11 |
"attention_dropout": 0.1,
|
12 |
"bos_token_id": 2,
|
13 |
+
"control_symbol_vocoder_offset": 4,
|
14 |
"conv_depthwise_kernel_size": 31,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
"decoder_attention_heads": 16,
|
16 |
"decoder_ffn_dim": 4096,
|
17 |
"decoder_layerdrop": 0.05,
|
|
|
23 |
"encoder_layerdrop": 0.05,
|
24 |
"encoder_layers": 12,
|
25 |
"eos_token_id": 3,
|
26 |
+
"feature_projection_input_dim": 160,
|
27 |
"hidden_size": 1024,
|
|
|
28 |
"initializer_range": 0.02,
|
29 |
"is_encoder_decoder": true,
|
30 |
"lang_embed_dim": 256,
|
31 |
"layer_norm_eps": 1e-05,
|
|
|
32 |
"leaky_relu_slope": 0.1,
|
33 |
"max_new_tokens": 256,
|
34 |
"max_position_embeddings": 4096,
|
35 |
"max_source_positions": 4096,
|
|
|
36 |
"model_type": "seamless_m4t",
|
37 |
"num_adapter_layers": 1,
|
38 |
"num_attention_heads": 16,
|
39 |
"num_conv_pos_embedding_groups": 16,
|
40 |
"num_conv_pos_embeddings": 128,
|
41 |
"num_hidden_layers": 12,
|
|
|
42 |
"pad_token_id": 0,
|
43 |
"position_embeddings_type": "relative",
|
44 |
"resblock_dilation_sizes": [
|
|
|
70 |
"speech_encoder_dropout": 0.0,
|
71 |
"speech_encoder_hidden_act": "swish",
|
72 |
"speech_encoder_intermediate_size": 4096,
|
73 |
+
"speech_encoder_layerdrop": 0.1,
|
74 |
"speech_encoder_layers": 12,
|
75 |
"spkr_embed_dim": 256,
|
76 |
"t2u_bos_token_id": 0,
|
|
|
83 |
"t2u_encoder_layers": 4,
|
84 |
"t2u_eos_token_id": 2,
|
85 |
"t2u_max_new_tokens": 1024,
|
86 |
+
"t2u_max_position_embeddings": 2048,
|
87 |
"t2u_num_langs": 38,
|
88 |
+
"t2u_offset_tgt_lang": 10005,
|
89 |
"t2u_pad_token_id": 1,
|
90 |
"torch_dtype": "float32",
|
91 |
"transformers_version": "4.33.0.dev0",
|