Hugging Face's logo

tahmaz
/
seamless-m4t-v2-large-azerbaijani

Safetensors
seamless_m4t_v2
Model card Files
xet
 Community
tahmaz commited on
Commit
2e47ab1
·
verified ·
1 Parent(s): 443347e

Incremental update: SeamlessM4Tv2 Large Speech-to-Text Azerbaijani

Browse files
Files changed (8) hide show
  1. .gitattributes +1 -0
  2. config.json +116 -0
  3. generation_config.json +0 -0
  4. model.safetensors +3 -0
  5. processor_config.json +113 -0
  6. tokenizer.json +3 -0
  7. tokenizer_config.json +119 -0
  8. training_args.bin +3 -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 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "SeamlessM4Tv2ForSpeechToText"
10
+ ],
11
+ "attention_dropout": 0.1,
12
+ "bos_token_id": 2,
13
+ "char_vocab_size": 10943,
14
+ "conv_depthwise_kernel_size": 31,
15
+ "decoder_attention_heads": 16,
16
+ "decoder_ffn_dim": 8192,
17
+ "decoder_layerdrop": 0.05,
18
+ "decoder_layers": 24,
19
+ "decoder_start_token_id": 3,
20
+ "dropout": 0.1,
21
+ "dtype": "bfloat16",
22
+ "encoder_attention_heads": 16,
23
+ "encoder_ffn_dim": 8192,
24
+ "encoder_layerdrop": 0.05,
25
+ "encoder_layers": 24,
26
+ "eos_token_id": 3,
27
+ "feature_projection_input_dim": 160,
28
+ "hidden_size": 1024,
29
+ "initializer_range": 0.02,
30
+ "is_encoder_decoder": true,
31
+ "lang_embed_dim": 256,
32
+ "layer_norm_eps": 1e-05,
33
+ "leaky_relu_slope": 0.1,
34
+ "left_max_position_embeddings": 64,
35
+ "max_new_tokens": 256,
36
+ "max_position_embeddings": 4096,
37
+ "model_type": "seamless_m4t_v2",
38
+ "num_adapter_layers": 1,
39
+ "pad_token_id": 0,
40
+ "position_embeddings_type": "relative_key",
41
+ "resblock_dilation_sizes": [
42
+ [
43
+ 1,
44
+ 3,
45
+ 5
46
+ ],
47
+ [
48
+ 1,
49
+ 3,
50
+ 5
51
+ ],
52
+ [
53
+ 1,
54
+ 3,
55
+ 5
56
+ ]
57
+ ],
58
+ "resblock_kernel_sizes": [
59
+ 3,
60
+ 7,
61
+ 11
62
+ ],
63
+ "right_max_position_embeddings": 8,
64
+ "sampling_rate": 16000,
65
+ "scale_embedding": true,
66
+ "speech_encoder_attention_heads": 16,
67
+ "speech_encoder_chunk_size": 20000,
68
+ "speech_encoder_dropout": 0.0,
69
+ "speech_encoder_hidden_act": "swish",
70
+ "speech_encoder_intermediate_size": 4096,
71
+ "speech_encoder_layerdrop": 0.1,
72
+ "speech_encoder_layers": 24,
73
+ "speech_encoder_left_chunk_num": 128,
74
+ "spkr_embed_dim": 256,
75
+ "t2u_bos_token_id": 0,
76
+ "t2u_decoder_attention_heads": 16,
77
+ "t2u_decoder_ffn_dim": 8192,
78
+ "t2u_decoder_layers": 6,
79
+ "t2u_encoder_attention_heads": 16,
80
+ "t2u_encoder_ffn_dim": 8192,
81
+ "t2u_encoder_layers": 6,
82
+ "t2u_eos_token_id": 2,
83
+ "t2u_max_position_embeddings": 4096,
84
+ "t2u_pad_token_id": 1,
85
+ "t2u_variance_pred_dropout": 0.5,
86
+ "t2u_variance_predictor_embed_dim": 1024,
87
+ "t2u_variance_predictor_hidden_dim": 256,
88
+ "t2u_variance_predictor_kernel_size": 3,
89
+ "t2u_vocab_size": 10082,
90
+ "tie_word_embeddings": true,
91
+ "transformers_version": "5.9.0",
92
+ "unit_embed_dim": 1280,
93
+ "unit_hifi_gan_vocab_size": 10000,
94
+ "upsample_initial_channel": 512,
95
+ "upsample_kernel_sizes": [
96
+ 11,
97
+ 8,
98
+ 8,
99
+ 4,
100
+ 4
101
+ ],
102
+ "upsample_rates": [
103
+ 5,
104
+ 4,
105
+ 4,
106
+ 2,
107
+ 2
108
+ ],
109
+ "use_cache": false,
110
+ "var_pred_dropout": 0.5,
111
+ "variance_predictor_kernel_size": 3,
112
+ "vocab_size": 256102,
113
+ "vocoder_num_langs": 36,
114
+ "vocoder_num_spkrs": 200,
115
+ "vocoder_offset": 4
116
+ }
generation_config.json ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f0b95fe9f49e607a6e7c24a3565da7a35cc9900b957886ce0b99d1814887e31
3
+ size 3003868616
processor_config.json ADDED
@@ -0,0 +1,113 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "feature_extractor": {
3
+ "feature_extractor_type": "SeamlessM4TFeatureExtractor",
4
+ "feature_size": 80,
5
+ "language_code": [
6
+ "__afr__",
7
+ "__amh__",
8
+ "__arb__",
9
+ "__ary__",
10
+ "__arz__",
11
+ "__asm__",
12
+ "__azj__",
13
+ "__bel__",
14
+ "__ben__",
15
+ "__bos__",
16
+ "__bul__",
17
+ "__cat__",
18
+ "__ceb__",
19
+ "__ces__",
20
+ "__ckb__",
21
+ "__cmn__",
22
+ "__cmn_Hant__",
23
+ "__cym__",
24
+ "__dan__",
25
+ "__deu__",
26
+ "__ell__",
27
+ "__eng__",
28
+ "__est__",
29
+ "__eus__",
30
+ "__fin__",
31
+ "__fra__",
32
+ "__fuv__",
33
+ "__gaz__",
34
+ "__gle__",
35
+ "__glg__",
36
+ "__guj__",
37
+ "__heb__",
38
+ "__hin__",
39
+ "__hrv__",
40
+ "__hun__",
41
+ "__hye__",
42
+ "__ibo__",
43
+ "__ind__",
44
+ "__isl__",
45
+ "__ita__",
46
+ "__jav__",
47
+ "__jpn__",
48
+ "__kan__",
49
+ "__kat__",
50
+ "__kaz__",
51
+ "__khk__",
52
+ "__khm__",
53
+ "__kir__",
54
+ "__kor__",
55
+ "__lao__",
56
+ "__lit__",
57
+ "__lug__",
58
+ "__luo__",
59
+ "__lvs__",
60
+ "__mai__",
61
+ "__mal__",
62
+ "__mar__",
63
+ "__mkd__",
64
+ "__mlt__",
65
+ "__mni__",
66
+ "__mya__",
67
+ "__nld__",
68
+ "__nno__",
69
+ "__nob__",
70
+ "__npi__",
71
+ "__nya__",
72
+ "__ory__",
73
+ "__pan__",
74
+ "__pbt__",
75
+ "__pes__",
76
+ "__pol__",
77
+ "__por__",
78
+ "__ron__",
79
+ "__rus__",
80
+ "__sat__",
81
+ "__slk__",
82
+ "__slv__",
83
+ "__sna__",
84
+ "__snd__",
85
+ "__som__",
86
+ "__spa__",
87
+ "__srp__",
88
+ "__swe__",
89
+ "__swh__",
90
+ "__tam__",
91
+ "__tel__",
92
+ "__tgk__",
93
+ "__tgl__",
94
+ "__tha__",
95
+ "__tur__",
96
+ "__ukr__",
97
+ "__urd__",
98
+ "__uzn__",
99
+ "__vie__",
100
+ "__yor__",
101
+ "__yue__",
102
+ "__zlm__",
103
+ "__zul__"
104
+ ],
105
+ "num_mel_bins": 80,
106
+ "padding_side": "right",
107
+ "padding_value": 0.0,
108
+ "return_attention_mask": true,
109
+ "sampling_rate": 16000,
110
+ "stride": 2
111
+ },
112
+ "processor_class": "SeamlessM4TProcessor"
113
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c77d1062bc72347c870653acb9e331a07cc240f3e1d53872a90343792ba4f61
3
+ size 30903093
tokenizer_config.json ADDED
@@ -0,0 +1,119 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "__afr__",
9
+ "__amh__",
10
+ "__arb__",
11
+ "__ary__",
12
+ "__arz__",
13
+ "__asm__",
14
+ "__azj__",
15
+ "__bel__",
16
+ "__ben__",
17
+ "__bos__",
18
+ "__bul__",
19
+ "__cat__",
20
+ "__ceb__",
21
+ "__ces__",
22
+ "__ckb__",
23
+ "__cmn__",
24
+ "__cmn_Hant__",
25
+ "__cym__",
26
+ "__dan__",
27
+ "__deu__",
28
+ "__ell__",
29
+ "__eng__",
30
+ "__est__",
31
+ "__eus__",
32
+ "__fin__",
33
+ "__fra__",
34
+ "__fuv__",
35
+ "__gaz__",
36
+ "__gle__",
37
+ "__glg__",
38
+ "__guj__",
39
+ "__heb__",
40
+ "__hin__",
41
+ "__hrv__",
42
+ "__hun__",
43
+ "__hye__",
44
+ "__ibo__",
45
+ "__ind__",
46
+ "__isl__",
47
+ "__ita__",
48
+ "__jav__",
49
+ "__jpn__",
50
+ "__kan__",
51
+ "__kat__",
52
+ "__kaz__",
53
+ "__khk__",
54
+ "__khm__",
55
+ "__kir__",
56
+ "__kor__",
57
+ "__lao__",
58
+ "__lit__",
59
+ "__lug__",
60
+ "__luo__",
61
+ "__lvs__",
62
+ "__mai__",
63
+ "__mal__",
64
+ "__mar__",
65
+ "__mkd__",
66
+ "__mlt__",
67
+ "__mni__",
68
+ "__mya__",
69
+ "__nld__",
70
+ "__nno__",
71
+ "__nob__",
72
+ "__npi__",
73
+ "__nya__",
74
+ "__ory__",
75
+ "__pan__",
76
+ "__pbt__",
77
+ "__pes__",
78
+ "__pol__",
79
+ "__por__",
80
+ "__ron__",
81
+ "__rus__",
82
+ "__sat__",
83
+ "__slk__",
84
+ "__slv__",
85
+ "__sna__",
86
+ "__snd__",
87
+ "__som__",
88
+ "__spa__",
89
+ "__srp__",
90
+ "__swe__",
91
+ "__swh__",
92
+ "__tam__",
93
+ "__tel__",
94
+ "__tgk__",
95
+ "__tgl__",
96
+ "__tha__",
97
+ "__tur__",
98
+ "__ukr__",
99
+ "__urd__",
100
+ "__uzn__",
101
+ "__vie__",
102
+ "__yor__",
103
+ "__yue__",
104
+ "__zlm__",
105
+ "__zul__"
106
+ ],
107
+ "is_local": true,
108
+ "keep_accents": null,
109
+ "local_files_only": true,
110
+ "model_max_length": 1000000000000000019884624838656,
111
+ "pad_token": "<pad>",
112
+ "processor_class": "SeamlessM4TProcessor",
113
+ "sep_token": "</s>",
114
+ "sp_model_kwargs": {},
115
+ "src_lang": "__eng__",
116
+ "tgt_lang": "__azj__",
117
+ "tokenizer_class": "SeamlessM4TTokenizer",
118
+ "unk_token": "<unk>"
119
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5440c54043383beb1ebf8ebc697ba7e8f0f1e59d7e03efaa9044991b8a90f57
3
+ size 4984