Hugging Face's logo

hf-tiny-v2
/
tiny-random-SeamlessM4TForTextToSpeech

Safetensors
seamless_m4t
Model card Files
xet
 Community
hf-transformers-bot commited on
Commit
715f59d
·
verified ·
1 Parent(s): 52bdd1c

Update tiny models for SeamlessM4TForTextToSpeech

Browse files
Files changed (7) hide show
  1. .gitattributes +1 -0
  2. config.json +119 -0
  3. generation_config.json +11 -0
  4. model.safetensors +3 -0
  5. preprocessor_config.json +214 -0
  6. tokenizer.json +3 -0
  7. tokenizer_config.json +227 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
config.json ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.0,
3
+ "activation_function": "relu",
4
+ "adaptor_dropout": 0.1,
5
+ "adaptor_kernel_size": 8,
6
+ "adaptor_stride": 8,
7
+ "add_adapter": true,
8
+ "architectures": [
9
+ "SeamlessM4TForTextToSpeech"
10
+ ],
11
+ "attention_dropout": 0.1,
12
+ "attention_probs_dropout_prob": 0.1,
13
+ "bos_token_id": 2,
14
+ "conv_depthwise_kernel_size": 31,
15
+ "decoder_attention_heads": 2,
16
+ "decoder_ffn_dim": 6,
17
+ "decoder_layerdrop": 0.05,
18
+ "decoder_layers": 2,
19
+ "decoder_start_token_id": 3,
20
+ "dropout": 0.1,
21
+ "dtype": "float32",
22
+ "encoder_attention_heads": 2,
23
+ "encoder_ffn_dim": 6,
24
+ "encoder_layerdrop": 0.05,
25
+ "encoder_layers": 2,
26
+ "eos_token_id": 3,
27
+ "feature_projection_input_dim": 160,
28
+ "hidden_act": "gelu",
29
+ "hidden_dropout_prob": 0.1,
30
+ "hidden_size": 6,
31
+ "initializer_range": 0.02,
32
+ "is_encoder_decoder": true,
33
+ "lang_embed_dim": 6,
34
+ "layer_norm_eps": 1e-05,
35
+ "leaky_relu_slope": 0.1,
36
+ "max_new_tokens": null,
37
+ "max_position_embeddings": 256,
38
+ "max_source_positions": 256,
39
+ "model_type": "seamless_m4t",
40
+ "num_adapter_layers": 1,
41
+ "num_conv_pos_embedding_groups": 16,
42
+ "num_conv_pos_embeddings": 8,
43
+ "pad_token_id": 0,
44
+ "position_embeddings_type": "relative",
45
+ "resblock_dilation_sizes": [
46
+ [
47
+ 1,
48
+ 3,
49
+ 5
50
+ ],
51
+ [
52
+ 1,
53
+ 3,
54
+ 5
55
+ ],
56
+ [
57
+ 1,
58
+ 3,
59
+ 5
60
+ ]
61
+ ],
62
+ "resblock_kernel_sizes": [
63
+ 3,
64
+ 7,
65
+ 11
66
+ ],
67
+ "rotary_embedding_base": 10000,
68
+ "sampling_rate": 16000,
69
+ "scale_embedding": true,
70
+ "speech_encoder_attention_heads": 2,
71
+ "speech_encoder_dropout": 0.0,
72
+ "speech_encoder_hidden_act": "swish",
73
+ "speech_encoder_intermediate_size": 6,
74
+ "speech_encoder_layerdrop": 0.1,
75
+ "speech_encoder_layers": 2,
76
+ "spkr_embed_dim": 6,
77
+ "t2u_bos_token_id": 0,
78
+ "t2u_decoder_attention_heads": 2,
79
+ "t2u_decoder_ffn_dim": 6,
80
+ "t2u_decoder_layers": 2,
81
+ "t2u_decoder_start_token_id": 2,
82
+ "t2u_encoder_attention_heads": 2,
83
+ "t2u_encoder_ffn_dim": 6,
84
+ "t2u_encoder_layers": 2,
85
+ "t2u_eos_token_id": 2,
86
+ "t2u_max_new_tokens": 25,
87
+ "t2u_max_position_embeddings": 2048,
88
+ "t2u_num_langs": 0,
89
+ "t2u_offset_tgt_lang": 0,
90
+ "t2u_pad_token_id": 1,
91
+ "t2u_vocab_size": 20,
92
+ "tie_word_embeddings": true,
93
+ "transformers_version": "5.16.0.dev0",
94
+ "unit_embed_dim": 25,
95
+ "unit_hifi_gan_vocab_size": 20,
96
+ "unit_hifigan_vocab_vise": 20,
97
+ "upsample_initial_channel": 32,
98
+ "upsample_kernel_sizes": [
99
+ 11,
100
+ 8,
101
+ 8,
102
+ 4,
103
+ 4
104
+ ],
105
+ "upsample_rates": [
106
+ 5,
107
+ 4,
108
+ 4,
109
+ 2,
110
+ 2
111
+ ],
112
+ "use_cache": true,
113
+ "var_pred_dropout": 0.5,
114
+ "variance_predictor_kernel_size": 3,
115
+ "vocab_size": 256203,
116
+ "vocoder_num_langs": 5,
117
+ "vocoder_num_spkrs": 5,
118
+ "vocoder_offset": 0
119
+ }
generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 2,
4
+ "decoder_start_token_id": 3,
5
+ "eos_token_id": 3,
6
+ "output_attentions": false,
7
+ "output_hidden_states": false,
8
+ "pad_token_id": 0,
9
+ "transformers_version": "5.16.0.dev0",
10
+ "use_cache": true
11
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:562bcc381db9ad138005a208ecfeb3643fbef1c686d32306b535ec192a52800a
3
+ size 6457260
preprocessor_config.json ADDED
@@ -0,0 +1,214 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "feature_extractor_type": "SeamlessM4TFeatureExtractor",
3
+ "feature_size": 80,
4
+ "language_code": [
5
+ "__ace__",
6
+ "__ace_Latn__",
7
+ "__acm__",
8
+ "__acq__",
9
+ "__aeb__",
10
+ "__afr__",
11
+ "__ajp__",
12
+ "__aka__",
13
+ "__amh__",
14
+ "__apc__",
15
+ "__arb__",
16
+ "__ars__",
17
+ "__ary__",
18
+ "__arz__",
19
+ "__asm__",
20
+ "__ast__",
21
+ "__awa__",
22
+ "__ayr__",
23
+ "__azb__",
24
+ "__azj__",
25
+ "__bak__",
26
+ "__bam__",
27
+ "__ban__",
28
+ "__bel__",
29
+ "__bem__",
30
+ "__ben__",
31
+ "__bho__",
32
+ "__bjn__",
33
+ "__bjn_Latn__",
34
+ "__bod__",
35
+ "__bos__",
36
+ "__bug__",
37
+ "__bul__",
38
+ "__cat__",
39
+ "__ceb__",
40
+ "__ces__",
41
+ "__cjk__",
42
+ "__ckb__",
43
+ "__crh__",
44
+ "__cym__",
45
+ "__dan__",
46
+ "__deu__",
47
+ "__dik__",
48
+ "__dyu__",
49
+ "__dzo__",
50
+ "__ell__",
51
+ "__eng__",
52
+ "__epo__",
53
+ "__est__",
54
+ "__eus__",
55
+ "__ewe__",
56
+ "__fao__",
57
+ "__pes__",
58
+ "__fij__",
59
+ "__fin__",
60
+ "__fon__",
61
+ "__fra__",
62
+ "__fur__",
63
+ "__fuv__",
64
+ "__gla__",
65
+ "__gle__",
66
+ "__glg__",
67
+ "__grn__",
68
+ "__guj__",
69
+ "__hat__",
70
+ "__hau__",
71
+ "__heb__",
72
+ "__hin__",
73
+ "__hne__",
74
+ "__hrv__",
75
+ "__hun__",
76
+ "__hye__",
77
+ "__ibo__",
78
+ "__ilo__",
79
+ "__ind__",
80
+ "__isl__",
81
+ "__ita__",
82
+ "__jav__",
83
+ "__jpn__",
84
+ "__kab__",
85
+ "__kac__",
86
+ "__kam__",
87
+ "__kan__",
88
+ "__kas__",
89
+ "__kas_Deva__",
90
+ "__kat__",
91
+ "__knc__",
92
+ "__knc_Latn__",
93
+ "__kaz__",
94
+ "__kbp__",
95
+ "__kea__",
96
+ "__khm__",
97
+ "__kik__",
98
+ "__kin__",
99
+ "__kir__",
100
+ "__kmb__",
101
+ "__kon__",
102
+ "__kor__",
103
+ "__kmr__",
104
+ "__lao__",
105
+ "__lvs__",
106
+ "__lij__",
107
+ "__lim__",
108
+ "__lin__",
109
+ "__lit__",
110
+ "__lmo__",
111
+ "__ltg__",
112
+ "__ltz__",
113
+ "__lua__",
114
+ "__lug__",
115
+ "__luo__",
116
+ "__lus__",
117
+ "__mag__",
118
+ "__mai__",
119
+ "__mal__",
120
+ "__mar__",
121
+ "__min__",
122
+ "__mkd__",
123
+ "__plt__",
124
+ "__mlt__",
125
+ "__mni__",
126
+ "__khk__",
127
+ "__mos__",
128
+ "__mri__",
129
+ "__zsm__",
130
+ "__mya__",
131
+ "__nld__",
132
+ "__nno__",
133
+ "__nob__",
134
+ "__npi__",
135
+ "__nso__",
136
+ "__nus__",
137
+ "__nya__",
138
+ "__oci__",
139
+ "__gaz__",
140
+ "__ory__",
141
+ "__pag__",
142
+ "__pan__",
143
+ "__pap__",
144
+ "__pol__",
145
+ "__por__",
146
+ "__prs__",
147
+ "__pbt__",
148
+ "__quy__",
149
+ "__ron__",
150
+ "__run__",
151
+ "__rus__",
152
+ "__sag__",
153
+ "__san__",
154
+ "__sat__",
155
+ "__scn__",
156
+ "__shn__",
157
+ "__sin__",
158
+ "__slk__",
159
+ "__slv__",
160
+ "__smo__",
161
+ "__sna__",
162
+ "__snd__",
163
+ "__som__",
164
+ "__sot__",
165
+ "__spa__",
166
+ "__als__",
167
+ "__srd__",
168
+ "__srp__",
169
+ "__ssw__",
170
+ "__sun__",
171
+ "__swe__",
172
+ "__swh__",
173
+ "__szl__",
174
+ "__tam__",
175
+ "__tat__",
176
+ "__tel__",
177
+ "__tgk__",
178
+ "__tgl__",
179
+ "__tha__",
180
+ "__tir__",
181
+ "__taq__",
182
+ "__taq_Tfng__",
183
+ "__tpi__",
184
+ "__tsn__",
185
+ "__tso__",
186
+ "__tuk__",
187
+ "__tum__",
188
+ "__tur__",
189
+ "__twi__",
190
+ "__tzm__",
191
+ "__uig__",
192
+ "__ukr__",
193
+ "__umb__",
194
+ "__urd__",
195
+ "__uzn__",
196
+ "__vec__",
197
+ "__vie__",
198
+ "__war__",
199
+ "__wol__",
200
+ "__xho__",
201
+ "__ydd__",
202
+ "__yor__",
203
+ "__yue__",
204
+ "__cmn__",
205
+ "__cmn_Hant__",
206
+ "__zul__"
207
+ ],
208
+ "num_mel_bins": 80,
209
+ "padding_side": "right",
210
+ "padding_value": 0.0,
211
+ "return_attention_mask": true,
212
+ "sampling_rate": 16000,
213
+ "stride": 2
214
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1b304a1bd353227b865490d83d16f28b6a6ce4be60607dd021f359e118063deb
3
+ size 31918020
tokenizer_config.json ADDED
@@ -0,0 +1,227 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "bos_token": "<s>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "extra_special_tokens": [
8
+ "<pad>",
9
+ "<unk>",
10
+ "<s>",
11
+ "</s>",
12
+ "__ace__",
13
+ "__ace_Latn__",
14
+ "__acm__",
15
+ "__acq__",
16
+ "__aeb__",
17
+ "__afr__",
18
+ "__ajp__",
19
+ "__aka__",
20
+ "__amh__",
21
+ "__apc__",
22
+ "__arb__",
23
+ "__ars__",
24
+ "__ary__",
25
+ "__arz__",
26
+ "__asm__",
27
+ "__ast__",
28
+ "__awa__",
29
+ "__ayr__",
30
+ "__azb__",
31
+ "__azj__",
32
+ "__bak__",
33
+ "__bam__",
34
+ "__ban__",
35
+ "__bel__",
36
+ "__bem__",
37
+ "__ben__",
38
+ "__bho__",
39
+ "__bjn__",
40
+ "__bjn_Latn__",
41
+ "__bod__",
42
+ "__bos__",
43
+ "__bug__",
44
+ "__bul__",
45
+ "__cat__",
46
+ "__ceb__",
47
+ "__ces__",
48
+ "__cjk__",
49
+ "__ckb__",
50
+ "__crh__",
51
+ "__cym__",
52
+ "__dan__",
53
+ "__deu__",
54
+ "__dik__",
55
+ "__dyu__",
56
+ "__dzo__",
57
+ "__ell__",
58
+ "__eng__",
59
+ "__epo__",
60
+ "__est__",
61
+ "__eus__",
62
+ "__ewe__",
63
+ "__fao__",
64
+ "__pes__",
65
+ "__fij__",
66
+ "__fin__",
67
+ "__fon__",
68
+ "__fra__",
69
+ "__fur__",
70
+ "__fuv__",
71
+ "__gla__",
72
+ "__gle__",
73
+ "__glg__",
74
+ "__grn__",
75
+ "__guj__",
76
+ "__hat__",
77
+ "__hau__",
78
+ "__heb__",
79
+ "__hin__",
80
+ "__hne__",
81
+ "__hrv__",
82
+ "__hun__",
83
+ "__hye__",
84
+ "__ibo__",
85
+ "__ilo__",
86
+ "__ind__",
87
+ "__isl__",
88
+ "__ita__",
89
+ "__jav__",
90
+ "__jpn__",
91
+ "__kab__",
92
+ "__kac__",
93
+ "__kam__",
94
+ "__kan__",
95
+ "__kas__",
96
+ "__kas_Deva__",
97
+ "__kat__",
98
+ "__knc__",
99
+ "__knc_Latn__",
100
+ "__kaz__",
101
+ "__kbp__",
102
+ "__kea__",
103
+ "__khm__",
104
+ "__kik__",
105
+ "__kin__",
106
+ "__kir__",
107
+ "__kmb__",
108
+ "__kon__",
109
+ "__kor__",
110
+ "__kmr__",
111
+ "__lao__",
112
+ "__lvs__",
113
+ "__lij__",
114
+ "__lim__",
115
+ "__lin__",
116
+ "__lit__",
117
+ "__lmo__",
118
+ "__ltg__",
119
+ "__ltz__",
120
+ "__lua__",
121
+ "__lug__",
122
+ "__luo__",
123
+ "__lus__",
124
+ "__mag__",
125
+ "__mai__",
126
+ "__mal__",
127
+ "__mar__",
128
+ "__min__",
129
+ "__mkd__",
130
+ "__plt__",
131
+ "__mlt__",
132
+ "__mni__",
133
+ "__khk__",
134
+ "__mos__",
135
+ "__mri__",
136
+ "__zsm__",
137
+ "__mya__",
138
+ "__nld__",
139
+ "__nno__",
140
+ "__nob__",
141
+ "__npi__",
142
+ "__nso__",
143
+ "__nus__",
144
+ "__nya__",
145
+ "__oci__",
146
+ "__gaz__",
147
+ "__ory__",
148
+ "__pag__",
149
+ "__pan__",
150
+ "__pap__",
151
+ "__pol__",
152
+ "__por__",
153
+ "__prs__",
154
+ "__pbt__",
155
+ "__quy__",
156
+ "__ron__",
157
+ "__run__",
158
+ "__rus__",
159
+ "__sag__",
160
+ "__san__",
161
+ "__sat__",
162
+ "__scn__",
163
+ "__shn__",
164
+ "__sin__",
165
+ "__slk__",
166
+ "__slv__",
167
+ "__smo__",
168
+ "__sna__",
169
+ "__snd__",
170
+ "__som__",
171
+ "__sot__",
172
+ "__spa__",
173
+ "__als__",
174
+ "__srd__",
175
+ "__srp__",
176
+ "__ssw__",
177
+ "__sun__",
178
+ "__swe__",
179
+ "__swh__",
180
+ "__szl__",
181
+ "__tam__",
182
+ "__tat__",
183
+ "__tel__",
184
+ "__tgk__",
185
+ "__tgl__",
186
+ "__tha__",
187
+ "__tir__",
188
+ "__taq__",
189
+ "__taq_Tfng__",
190
+ "__tpi__",
191
+ "__tsn__",
192
+ "__tso__",
193
+ "__tuk__",
194
+ "__tum__",
195
+ "__tur__",
196
+ "__twi__",
197
+ "__tzm__",
198
+ "__uig__",
199
+ "__ukr__",
200
+ "__umb__",
201
+ "__urd__",
202
+ "__uzn__",
203
+ "__vec__",
204
+ "__vie__",
205
+ "__war__",
206
+ "__wol__",
207
+ "__xho__",
208
+ "__ydd__",
209
+ "__yor__",
210
+ "__yue__",
211
+ "__cmn__",
212
+ "__cmn_Hant__",
213
+ "__zul__"
214
+ ],
215
+ "is_local": true,
216
+ "keep_accents": null,
217
+ "local_files_only": false,
218
+ "model_max_length": 256,
219
+ "pad_token": "<pad>",
220
+ "processor_class": "SeamlessM4TProcessor",
221
+ "sep_token": "</s>",
222
+ "sp_model_kwargs": {},
223
+ "src_lang": "__eng__",
224
+ "tgt_lang": "__fra__",
225
+ "tokenizer_class": "SeamlessM4TTokenizer",
226
+ "unk_token": "<unk>"
227
+ }