ylacombe HF staff commited on
Commit
44accf6
1 Parent(s): b8199f2

Upload ParlerTTSForConditionalGeneration

Browse files
config.json ADDED
@@ -0,0 +1,282 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/fsx/yoach/tmp/artefacts/training-50K-large-without-accents-v2-80k-steps/",
3
+ "architectures": [
4
+ "ParlerTTSForConditionalGeneration"
5
+ ],
6
+ "audio_encoder": {
7
+ "_name_or_path": "parler-tts/dac_44khZ_8kbps",
8
+ "add_cross_attention": false,
9
+ "architectures": [
10
+ "DACModel"
11
+ ],
12
+ "bad_words_ids": null,
13
+ "begin_suppress_tokens": null,
14
+ "bos_token_id": null,
15
+ "chunk_size_feed_forward": 0,
16
+ "codebook_size": 1024,
17
+ "cross_attention_hidden_size": null,
18
+ "decoder_start_token_id": null,
19
+ "diversity_penalty": 0.0,
20
+ "do_sample": false,
21
+ "early_stopping": false,
22
+ "encoder_no_repeat_ngram_size": 0,
23
+ "eos_token_id": null,
24
+ "exponential_decay_length_penalty": null,
25
+ "finetuning_task": null,
26
+ "forced_bos_token_id": null,
27
+ "forced_eos_token_id": null,
28
+ "frame_rate": 86,
29
+ "id2label": {
30
+ "0": "LABEL_0",
31
+ "1": "LABEL_1"
32
+ },
33
+ "is_decoder": false,
34
+ "is_encoder_decoder": false,
35
+ "label2id": {
36
+ "LABEL_0": 0,
37
+ "LABEL_1": 1
38
+ },
39
+ "latent_dim": 1024,
40
+ "length_penalty": 1.0,
41
+ "max_length": 20,
42
+ "min_length": 0,
43
+ "model_bitrate": 8,
44
+ "model_type": "dac",
45
+ "no_repeat_ngram_size": 0,
46
+ "num_beam_groups": 1,
47
+ "num_beams": 1,
48
+ "num_codebooks": 9,
49
+ "num_return_sequences": 1,
50
+ "output_attentions": false,
51
+ "output_hidden_states": false,
52
+ "output_scores": false,
53
+ "pad_token_id": null,
54
+ "prefix": null,
55
+ "problem_type": null,
56
+ "pruned_heads": {},
57
+ "remove_invalid_values": false,
58
+ "repetition_penalty": 1.0,
59
+ "return_dict": true,
60
+ "return_dict_in_generate": false,
61
+ "sampling_rate": 44100,
62
+ "sep_token_id": null,
63
+ "suppress_tokens": null,
64
+ "task_specific_params": null,
65
+ "temperature": 1.0,
66
+ "tf_legacy_loss": false,
67
+ "tie_encoder_decoder": false,
68
+ "tie_word_embeddings": true,
69
+ "tokenizer_class": null,
70
+ "top_k": 50,
71
+ "top_p": 1.0,
72
+ "torch_dtype": "float32",
73
+ "torchscript": false,
74
+ "typical_p": 1.0,
75
+ "use_bfloat16": false
76
+ },
77
+ "decoder": {
78
+ "_name_or_path": "/fsx/yoach/tmp/artefacts/parler-tts-large-v2/decoder",
79
+ "activation_dropout": 0.0,
80
+ "activation_function": "gelu",
81
+ "add_cross_attention": true,
82
+ "architectures": [
83
+ "ParlerTTSForCausalLM"
84
+ ],
85
+ "attention_dropout": 0.0,
86
+ "bad_words_ids": null,
87
+ "begin_suppress_tokens": null,
88
+ "bos_token_id": 1025,
89
+ "chunk_size_feed_forward": 0,
90
+ "cross_attention_hidden_size": null,
91
+ "cross_attention_implementation_strategy": "always_sdpa",
92
+ "decoder_start_token_id": null,
93
+ "diversity_penalty": 0.0,
94
+ "do_sample": false,
95
+ "dropout": 0.1,
96
+ "early_stopping": false,
97
+ "encoder_no_repeat_ngram_size": 0,
98
+ "eos_token_id": 1024,
99
+ "exponential_decay_length_penalty": null,
100
+ "ffn_dim": 6144,
101
+ "finetuning_task": null,
102
+ "forced_bos_token_id": null,
103
+ "forced_eos_token_id": null,
104
+ "hidden_size": 1536,
105
+ "id2label": {
106
+ "0": "LABEL_0",
107
+ "1": "LABEL_1"
108
+ },
109
+ "initializer_factor": 0.02,
110
+ "is_decoder": true,
111
+ "is_encoder_decoder": false,
112
+ "label2id": {
113
+ "LABEL_0": 0,
114
+ "LABEL_1": 1
115
+ },
116
+ "layerdrop": 0.0,
117
+ "length_penalty": 1.0,
118
+ "max_length": 20,
119
+ "max_position_embeddings": 4096,
120
+ "min_length": 0,
121
+ "model_type": "parler_tts_decoder",
122
+ "no_repeat_ngram_size": 0,
123
+ "num_attention_heads": 24,
124
+ "num_beam_groups": 1,
125
+ "num_beams": 1,
126
+ "num_codebooks": 9,
127
+ "num_cross_attention_key_value_heads": 24,
128
+ "num_hidden_layers": 32,
129
+ "num_key_value_heads": 8,
130
+ "num_return_sequences": 1,
131
+ "output_attentions": false,
132
+ "output_hidden_states": false,
133
+ "output_scores": false,
134
+ "pad_token_id": 1024,
135
+ "prefix": null,
136
+ "problem_type": null,
137
+ "pruned_heads": {},
138
+ "remove_invalid_values": false,
139
+ "repetition_penalty": 1.0,
140
+ "return_dict": true,
141
+ "return_dict_in_generate": false,
142
+ "rope_embeddings": false,
143
+ "rope_theta": 10000.0,
144
+ "scale_embedding": false,
145
+ "sep_token_id": null,
146
+ "suppress_tokens": null,
147
+ "task_specific_params": null,
148
+ "temperature": 1.0,
149
+ "tf_legacy_loss": false,
150
+ "tie_encoder_decoder": false,
151
+ "tie_word_embeddings": false,
152
+ "tokenizer_class": null,
153
+ "top_k": 50,
154
+ "top_p": 1.0,
155
+ "torch_dtype": "float32",
156
+ "torchscript": false,
157
+ "typical_p": 1.0,
158
+ "use_bfloat16": false,
159
+ "use_cache": true,
160
+ "vocab_size": 1088
161
+ },
162
+ "decoder_start_token_id": 1025,
163
+ "is_encoder_decoder": true,
164
+ "model_type": "parler_tts",
165
+ "pad_token_id": 1024,
166
+ "prompt_cross_attention": true,
167
+ "text_encoder": {
168
+ "_name_or_path": "google/flan-t5-xl",
169
+ "add_cross_attention": false,
170
+ "architectures": [
171
+ "T5ForConditionalGeneration"
172
+ ],
173
+ "bad_words_ids": null,
174
+ "begin_suppress_tokens": null,
175
+ "bos_token_id": null,
176
+ "chunk_size_feed_forward": 0,
177
+ "classifier_dropout": 0.0,
178
+ "cross_attention_hidden_size": null,
179
+ "d_ff": 5120,
180
+ "d_kv": 64,
181
+ "d_model": 2048,
182
+ "decoder_start_token_id": 0,
183
+ "dense_act_fn": "gelu_new",
184
+ "diversity_penalty": 0.0,
185
+ "do_sample": false,
186
+ "dropout_rate": 0.1,
187
+ "early_stopping": false,
188
+ "encoder_no_repeat_ngram_size": 0,
189
+ "eos_token_id": 1,
190
+ "exponential_decay_length_penalty": null,
191
+ "feed_forward_proj": "gated-gelu",
192
+ "finetuning_task": null,
193
+ "forced_bos_token_id": null,
194
+ "forced_eos_token_id": null,
195
+ "id2label": {
196
+ "0": "LABEL_0",
197
+ "1": "LABEL_1"
198
+ },
199
+ "initializer_factor": 1.0,
200
+ "is_decoder": false,
201
+ "is_encoder_decoder": true,
202
+ "is_gated_act": true,
203
+ "label2id": {
204
+ "LABEL_0": 0,
205
+ "LABEL_1": 1
206
+ },
207
+ "layer_norm_epsilon": 1e-06,
208
+ "length_penalty": 1.0,
209
+ "max_length": 20,
210
+ "min_length": 0,
211
+ "model_type": "t5",
212
+ "n_positions": 512,
213
+ "no_repeat_ngram_size": 0,
214
+ "num_beam_groups": 1,
215
+ "num_beams": 1,
216
+ "num_decoder_layers": 24,
217
+ "num_heads": 32,
218
+ "num_layers": 24,
219
+ "num_return_sequences": 1,
220
+ "output_attentions": false,
221
+ "output_hidden_states": false,
222
+ "output_past": true,
223
+ "output_scores": false,
224
+ "pad_token_id": 0,
225
+ "prefix": null,
226
+ "problem_type": null,
227
+ "pruned_heads": {},
228
+ "relative_attention_max_distance": 128,
229
+ "relative_attention_num_buckets": 32,
230
+ "remove_invalid_values": false,
231
+ "repetition_penalty": 1.0,
232
+ "return_dict": true,
233
+ "return_dict_in_generate": false,
234
+ "sep_token_id": null,
235
+ "suppress_tokens": null,
236
+ "task_specific_params": {
237
+ "summarization": {
238
+ "early_stopping": true,
239
+ "length_penalty": 2.0,
240
+ "max_length": 200,
241
+ "min_length": 30,
242
+ "no_repeat_ngram_size": 3,
243
+ "num_beams": 4,
244
+ "prefix": "summarize: "
245
+ },
246
+ "translation_en_to_de": {
247
+ "early_stopping": true,
248
+ "max_length": 300,
249
+ "num_beams": 4,
250
+ "prefix": "translate English to German: "
251
+ },
252
+ "translation_en_to_fr": {
253
+ "early_stopping": true,
254
+ "max_length": 300,
255
+ "num_beams": 4,
256
+ "prefix": "translate English to French: "
257
+ },
258
+ "translation_en_to_ro": {
259
+ "early_stopping": true,
260
+ "max_length": 300,
261
+ "num_beams": 4,
262
+ "prefix": "translate English to Romanian: "
263
+ }
264
+ },
265
+ "temperature": 1.0,
266
+ "tf_legacy_loss": false,
267
+ "tie_encoder_decoder": false,
268
+ "tie_word_embeddings": false,
269
+ "tokenizer_class": null,
270
+ "top_k": 50,
271
+ "top_p": 1.0,
272
+ "torch_dtype": "float32",
273
+ "torchscript": false,
274
+ "typical_p": 1.0,
275
+ "use_bfloat16": false,
276
+ "use_cache": true,
277
+ "vocab_size": 32128
278
+ },
279
+ "torch_dtype": "float32",
280
+ "transformers_version": "4.40.2",
281
+ "vocab_size": 32128
282
+ }
generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1025,
4
+ "decoder_start_token_id": 1025,
5
+ "do_sample": true,
6
+ "eos_token_id": 1024,
7
+ "guidance_scale": 1,
8
+ "max_length": 2580,
9
+ "pad_token_id": 1024,
10
+ "transformers_version": "4.40.2"
11
+ }
model-00001-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c30d2151a8a9c3343b6998eeec019b46db9d17b84ef234d98ea6723941c2851c
3
+ size 4984365952
model-00002-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfa7124014485313228b170ff6cd42bc6b1f9476bb4e833e15a0dd378e0e5e71
3
+ size 4805000272
model-00003-of-00003.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b866c89a85f45d4f4e045f72f1b9f2760de1d72549d7de0dbc73d2bb20df1685
3
+ size 222560488
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff