Pis-py commited on
Commit
27bea24
·
verified ·
1 Parent(s): a625c16

Upload 10 files

Browse files
.gitattributes CHANGED
@@ -1,35 +1,36 @@
1
- *.7z filter=lfs diff=lfs merge=lfs -text
2
- *.arrow filter=lfs diff=lfs merge=lfs -text
3
- *.bin filter=lfs diff=lfs merge=lfs -text
4
- *.bz2 filter=lfs diff=lfs merge=lfs -text
5
- *.ckpt filter=lfs diff=lfs merge=lfs -text
6
- *.ftz filter=lfs diff=lfs merge=lfs -text
7
- *.gz filter=lfs diff=lfs merge=lfs -text
8
- *.h5 filter=lfs diff=lfs merge=lfs -text
9
- *.joblib filter=lfs diff=lfs merge=lfs -text
10
- *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
- *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
- *.model filter=lfs diff=lfs merge=lfs -text
13
- *.msgpack filter=lfs diff=lfs merge=lfs -text
14
- *.npy filter=lfs diff=lfs merge=lfs -text
15
- *.npz filter=lfs diff=lfs merge=lfs -text
16
- *.onnx filter=lfs diff=lfs merge=lfs -text
17
- *.ot filter=lfs diff=lfs merge=lfs -text
18
- *.parquet filter=lfs diff=lfs merge=lfs -text
19
- *.pb filter=lfs diff=lfs merge=lfs -text
20
- *.pickle filter=lfs diff=lfs merge=lfs -text
21
- *.pkl filter=lfs diff=lfs merge=lfs -text
22
- *.pt filter=lfs diff=lfs merge=lfs -text
23
- *.pth filter=lfs diff=lfs merge=lfs -text
24
- *.rar filter=lfs diff=lfs merge=lfs -text
25
- *.safetensors filter=lfs diff=lfs merge=lfs -text
26
- saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
- *.tar.* filter=lfs diff=lfs merge=lfs -text
28
- *.tar filter=lfs diff=lfs merge=lfs -text
29
- *.tflite filter=lfs diff=lfs merge=lfs -text
30
- *.tgz filter=lfs diff=lfs merge=lfs -text
31
- *.wasm filter=lfs diff=lfs merge=lfs -text
32
- *.xz 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
 
 
1
+ *.7z filter=lfs diff=lfs merge=lfs -text
2
+ *.arrow filter=lfs diff=lfs merge=lfs -text
3
+ *.bin filter=lfs diff=lfs merge=lfs -text
4
+ *.bz2 filter=lfs diff=lfs merge=lfs -text
5
+ *.ckpt filter=lfs diff=lfs merge=lfs -text
6
+ *.ftz filter=lfs diff=lfs merge=lfs -text
7
+ *.gz filter=lfs diff=lfs merge=lfs -text
8
+ *.h5 filter=lfs diff=lfs merge=lfs -text
9
+ *.joblib filter=lfs diff=lfs merge=lfs -text
10
+ *.lfs.* filter=lfs diff=lfs merge=lfs -text
11
+ *.mlmodel filter=lfs diff=lfs merge=lfs -text
12
+ *.model filter=lfs diff=lfs merge=lfs -text
13
+ *.msgpack filter=lfs diff=lfs merge=lfs -text
14
+ *.npy filter=lfs diff=lfs merge=lfs -text
15
+ *.npz filter=lfs diff=lfs merge=lfs -text
16
+ *.onnx filter=lfs diff=lfs merge=lfs -text
17
+ *.ot filter=lfs diff=lfs merge=lfs -text
18
+ *.parquet filter=lfs diff=lfs merge=lfs -text
19
+ *.pb filter=lfs diff=lfs merge=lfs -text
20
+ *.pickle filter=lfs diff=lfs merge=lfs -text
21
+ *.pkl filter=lfs diff=lfs merge=lfs -text
22
+ *.pt filter=lfs diff=lfs merge=lfs -text
23
+ *.pth filter=lfs diff=lfs merge=lfs -text
24
+ *.rar filter=lfs diff=lfs merge=lfs -text
25
+ *.safetensors filter=lfs diff=lfs merge=lfs -text
26
+ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
+ *.tar.* filter=lfs diff=lfs merge=lfs -text
28
+ *.tar filter=lfs diff=lfs merge=lfs -text
29
+ *.tflite filter=lfs diff=lfs merge=lfs -text
30
+ *.tgz filter=lfs diff=lfs merge=lfs -text
31
+ *.wasm filter=lfs diff=lfs merge=lfs -text
32
+ *.xz 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
README.md ADDED
@@ -0,0 +1,199 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ language:
5
+ - wo
6
+ base_model:
7
+ - parler-tts/parler-tts-mini-multilingual-v1.1
8
+ pipeline_tag: text-to-speech
9
+ ---
10
+ <img src="https://huggingface.co/spaces/CONCREE/Adia_TTS/resolve/main/adia.png" alt=" Adia TTS Logo" width="800" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
11
+
12
+ # Adia_TTS Wolof
13
+
14
+ <a target="_blank" href="https://huggingface.co/spaces/CONCREE/Adia_TTS">
15
+ <img src="https://huggingface.co/datasets/huggingface/badges/raw/main/open-in-hf-spaces-sm.svg" alt="Open in HuggingFace"/>
16
+ </a>
17
+
18
+ # ADIA_TTS : Modèle de synthèse vocale en Wolof
19
+
20
+ ## Table des matières
21
+ 1. Introduction
22
+ 2. Caractéristiques techniques
23
+ 3. Installation
24
+ 4. Utilisation
25
+ 5. Configuration avancée
26
+ 6. Exemples
27
+ 7. Limitations
28
+ 8. Références
29
+ 9. Licence
30
+
31
+ ## Introduction
32
+ ADIA_TTS est un modèle de synthèse vocale (Text-to-Speech) open source en wolof, développé par CONCREE. Basé sur le modèle parler-tts-mini-multilingual-v1.1, il représente une avancée significative dans la synthèse vocale pour la langue wolof.
33
+
34
+ ## Points clés
35
+ * Entraîné sur 40 heures de données vocales en wolof
36
+ * Affiné pendant 100 epochs (~168 heures d'entraînement)
37
+ * Qualité vocale naturelle et fluide
38
+ * Voix unique avec contrôle des caractéristiques vocales via description
39
+
40
+
41
+ ## Caractéristiques techniques
42
+
43
+ ### Spécifications du modèle
44
+
45
+ * Architecture : parler-tts-mini-multilingual-v1.1
46
+ * Taille du modèle : 1.88 GB
47
+ * Format du modèle : PyTorch
48
+ * Fréquence d'échantillonnage : 24kHz
49
+ * Encodage audio : 16-bit PCM
50
+
51
+ ### Performance
52
+
53
+ * Temps moyen d'inférence : secondes/phrase (CPU), 20 secondes/phrase (GPU)
54
+ * Consommation mémoire : 3.9 GB (RAM minimale recommandée)
55
+
56
+ ## Installation
57
+ ### Prérequis
58
+ Python 3.8 ou supérieur
59
+ PyTorch 2.0 ou supérieur
60
+ CUDA (obligatoire, pour accélération GPU)
61
+ ```sh
62
+ pip install git+https://github.com/huggingface/parler-tts.git
63
+ ```
64
+
65
+ ## Utilisation
66
+ ## Exemple de base
67
+
68
+ ```py
69
+ import torch
70
+ from parler_tts import ParlerTTSForConditionalGeneration
71
+ from transformers import AutoTokenizer
72
+ import soundfile as sf
73
+
74
+ device = "cuda:0" if torch.cuda.is_available() else "cpu"
75
+
76
+ # Chargement du modèle
77
+ model = ParlerTTSForConditionalGeneration.from_pretrained("CONCREE/Adia_TTS").to(device)
78
+ tokenizer = AutoTokenizer.from_pretrained("CONCREE/Adia_TTS")
79
+
80
+ # Texte en wolof à synthétiser
81
+ text = "Entreprenariat ci Senegal dafa am solo lool ci yokkuteg koom-koom, di gëna yokk liggéey ak indi gis-gis yu bees ci dëkk bi."
82
+
83
+ # Description du style vocal
84
+ description = "A clear and educational voice, with a flow adapted to learning"
85
+
86
+ # Génération
87
+ input_ids = tokenizer(description, return_tensors="pt").input_ids.to(device)
88
+ prompt_ids = tokenizer(text, return_tensors="pt").input_ids.to(device)
89
+
90
+ audio = model.generate(
91
+ input_ids=input_ids,
92
+ prompt_input_ids=prompt_ids,
93
+ )
94
+
95
+ # Sauvegarde
96
+ sf.write("output.wav", audio.cpu().numpy().squeeze(), model.config.sampling_rate)
97
+ ```
98
+ <audio controls>
99
+ <source src="https://huggingface.co/spaces/CONCREE/Adia_TTS/resolve/main/snt.wav" type="audio/wav">
100
+ </audio>
101
+
102
+ ## Configuration avancée
103
+ ## Paramètres de génération
104
+
105
+ ```py
106
+ generation_config = {
107
+ "temperature": 0.8, # Contrôle la variabilité de la sortie
108
+ "max_new_tokens": 1000, # Longueur maximale de la séquence générée
109
+ "do_sample": True, # Active l'échantillonnage aléatoire
110
+ "top_k": 50, # Limite le nombre de tokens considérés
111
+ "repetition_penalty": 1.2, # Pénalise la répétition de tokens
112
+ }
113
+
114
+ audio = model.generate(
115
+ input_ids=input_ids,
116
+ prompt_input_ids=prompt_ids,
117
+ **generation_config
118
+ )
119
+ ```
120
+
121
+ <audio controls>
122
+ <source src="https://huggingface.co/spaces/CONCREE/Adia_TTS/resolve/main/avct.wav" type="audio/wav">
123
+ </audio>
124
+
125
+ ## Styles vocaux
126
+ ### Exemples de descriptions pour différents styles :
127
+
128
+ 1. ### Voix naturelle
129
+
130
+ ```py
131
+ description = "A warm and natural voice, with a conversational flow"
132
+ ```
133
+ <audio controls>
134
+ <source src="https://huggingface.co/spaces/CONCREE/Adia_TTS/resolve/main/styl1.wav" type="audio/wav">
135
+ </audio>
136
+
137
+ 2. ### Voix professionnelle
138
+
139
+ ```py
140
+ description = "A professional, clear and composed voice, perfect for formal presentations"
141
+ ```
142
+ <audio controls>
143
+ <source src="https://huggingface.co/spaces/CONCREE/Adia_TTS/resolve/main/styl2.wav" type="audio/wav">
144
+ </audio>
145
+
146
+ 3. ### Voix pour l'éducation
147
+ ```py
148
+ description = "A clear and educational voice, with a flow adapted to learning"
149
+ ```
150
+ <audio controls>
151
+ <source src="https://huggingface.co/spaces/CONCREE/Adia_TTS/resolve/main/styl3.wav" type="audio/wav">
152
+ </audio>
153
+
154
+ ## Limitations
155
+
156
+ * Performance réduite sur les phrases très longues
157
+ * Gestion limitée des nombres et des dates
158
+ * Temps de chargement initial du modèle relativement plus long
159
+ * Le modèle est limité à 200 caractères maximum par inférence sans segmentation. Pour les textes plus longs, une segmentation manuelle est nécessaire.
160
+ * La qualité des transitions entre segments peut varier selon la méthode de segmentation choisie
161
+ * Il est recommandé de segmenter le texte aux frontières naturelles (phrases, paragraphes) pour de meilleurs résultats
162
+ ## Références
163
+
164
+ ```
165
+ @misc{CONCREE-2024-Adia_TTS,
166
+ author = {CONCREE},
167
+ title = {Adia_TTS},
168
+ year = {2025},
169
+ publisher = {Hugging Face},
170
+ journal = {Hugging Face repository},
171
+ howpublished = {\url{https://huggingface.co/CONCREE/Adia_TTS}}
172
+ }
173
+ ```
174
+
175
+ ```
176
+ @misc{lyth2024natural,
177
+ title={Natural language guidance of high-fidelity text-to-speech with synthetic annotations},
178
+ author={Dan Lyth and Simon King},
179
+ year={2024},
180
+ eprint={2402.01912},
181
+ archivePrefix={arXiv},
182
+ primaryClass={cs.SD}
183
+ }
184
+ ```
185
+
186
+ ## Licence
187
+ Ce projet est sous licence Apache 2.0. Voir le fichier LICENSE pour plus de détails.
188
+
189
+ ## Conditions d'utilisation
190
+
191
+ * Les utilisateurs s'engagent à utiliser le modèle d'une manière qui respecte la langue wolof et la culture sénégalaise.
192
+ * Nous encourageons l'utilisation de ce modèle pour développer des solutions qui améliorent l'accessibilité numérique pour les locuteurs wolof et contribuent à réduire la fracture numérique. Les projets visant l'inclusion numérique sont particulièrement bienvenus.
193
+ * Toute utilisation du modèle doit mentionner CONCREE comme créateur original. Les utilisateurs sont fortement encouragés à partager leurs améliorations avec la communauté.
194
+ * L'utilisation commerciale est autorisée selon les termes de la licence Apache 2.0.
195
+
196
+ ## Contact
197
+ Pour toute question ou support :
198
+
199
+ Email : ai@concree.com
config.json ADDED
@@ -0,0 +1,277 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/content/drive/MyDrive/output_dir_training",
3
+ "architectures": [
4
+ "ParlerTTSForConditionalGeneration"
5
+ ],
6
+ "audio_encoder": {
7
+ "_attn_implementation_autoset": false,
8
+ "_name_or_path": "ylacombe/dac_44khz",
9
+ "add_cross_attention": false,
10
+ "architectures": [
11
+ "DacModel"
12
+ ],
13
+ "bad_words_ids": null,
14
+ "begin_suppress_tokens": null,
15
+ "bos_token_id": null,
16
+ "chunk_size_feed_forward": 0,
17
+ "codebook_dim": 8,
18
+ "codebook_loss_weight": 1.0,
19
+ "codebook_size": 1024,
20
+ "commitment_loss_weight": 0.25,
21
+ "cross_attention_hidden_size": null,
22
+ "decoder_hidden_size": 1536,
23
+ "decoder_start_token_id": null,
24
+ "diversity_penalty": 0.0,
25
+ "do_sample": false,
26
+ "downsampling_ratios": [
27
+ 2,
28
+ 4,
29
+ 8,
30
+ 8
31
+ ],
32
+ "early_stopping": false,
33
+ "encoder_hidden_size": 64,
34
+ "encoder_no_repeat_ngram_size": 0,
35
+ "eos_token_id": null,
36
+ "exponential_decay_length_penalty": null,
37
+ "finetuning_task": null,
38
+ "forced_bos_token_id": null,
39
+ "forced_eos_token_id": null,
40
+ "hidden_size": 1024,
41
+ "hop_length": 512,
42
+ "id2label": {
43
+ "0": "LABEL_0",
44
+ "1": "LABEL_1"
45
+ },
46
+ "is_decoder": false,
47
+ "is_encoder_decoder": false,
48
+ "label2id": {
49
+ "LABEL_0": 0,
50
+ "LABEL_1": 1
51
+ },
52
+ "length_penalty": 1.0,
53
+ "max_length": 20,
54
+ "min_length": 0,
55
+ "model_type": "dac",
56
+ "n_codebooks": 9,
57
+ "no_repeat_ngram_size": 0,
58
+ "num_beam_groups": 1,
59
+ "num_beams": 1,
60
+ "num_return_sequences": 1,
61
+ "output_attentions": false,
62
+ "output_hidden_states": false,
63
+ "output_scores": false,
64
+ "pad_token_id": null,
65
+ "prefix": null,
66
+ "problem_type": null,
67
+ "pruned_heads": {},
68
+ "quantizer_dropout": 0.0,
69
+ "remove_invalid_values": false,
70
+ "repetition_penalty": 1.0,
71
+ "return_dict": true,
72
+ "return_dict_in_generate": false,
73
+ "sampling_rate": 44100,
74
+ "sep_token_id": null,
75
+ "suppress_tokens": null,
76
+ "task_specific_params": null,
77
+ "temperature": 1.0,
78
+ "tf_legacy_loss": false,
79
+ "tie_encoder_decoder": false,
80
+ "tie_word_embeddings": true,
81
+ "tokenizer_class": null,
82
+ "top_k": 50,
83
+ "top_p": 1.0,
84
+ "torch_dtype": "float32",
85
+ "torchscript": false,
86
+ "typical_p": 1.0,
87
+ "upsampling_ratios": [
88
+ 8,
89
+ 8,
90
+ 4,
91
+ 2
92
+ ],
93
+ "use_bfloat16": false
94
+ },
95
+ "decoder": {
96
+ "_attn_implementation_autoset": false,
97
+ "_name_or_path": "/fsx/yoach/tmp/artefacts/parler-tts-mini-v2-empty/decoder",
98
+ "activation_dropout": 0.0,
99
+ "activation_function": "gelu",
100
+ "add_cross_attention": true,
101
+ "architectures": [
102
+ "ParlerTTSForCausalLM"
103
+ ],
104
+ "attention_dropout": 0.0,
105
+ "bad_words_ids": null,
106
+ "begin_suppress_tokens": null,
107
+ "bos_token_id": 1025,
108
+ "chunk_size_feed_forward": 0,
109
+ "codebook_weights": null,
110
+ "cross_attention_hidden_size": null,
111
+ "cross_attention_implementation_strategy": null,
112
+ "decoder_start_token_id": null,
113
+ "delay_strategy": "delay",
114
+ "diversity_penalty": 0.0,
115
+ "do_sample": false,
116
+ "dropout": 0.1,
117
+ "early_stopping": false,
118
+ "encoder_no_repeat_ngram_size": 0,
119
+ "eos_token_id": 1024,
120
+ "exponential_decay_length_penalty": null,
121
+ "ffn_dim": 4096,
122
+ "finetuning_task": null,
123
+ "forced_bos_token_id": null,
124
+ "forced_eos_token_id": null,
125
+ "hidden_size": 1024,
126
+ "id2label": {
127
+ "0": "LABEL_0",
128
+ "1": "LABEL_1"
129
+ },
130
+ "initializer_factor": 0.02,
131
+ "is_decoder": true,
132
+ "is_encoder_decoder": false,
133
+ "label2id": {
134
+ "LABEL_0": 0,
135
+ "LABEL_1": 1
136
+ },
137
+ "layerdrop": 0.0,
138
+ "length_penalty": 1.0,
139
+ "max_length": 20,
140
+ "max_position_embeddings": 4096,
141
+ "min_length": 0,
142
+ "model_type": "parler_tts_decoder",
143
+ "no_repeat_ngram_size": 0,
144
+ "num_attention_heads": 16,
145
+ "num_beam_groups": 1,
146
+ "num_beams": 1,
147
+ "num_codebooks": 9,
148
+ "num_cross_attention_key_value_heads": 16,
149
+ "num_hidden_layers": 24,
150
+ "num_key_value_heads": 16,
151
+ "num_return_sequences": 1,
152
+ "output_attentions": false,
153
+ "output_hidden_states": false,
154
+ "output_scores": false,
155
+ "pad_token_id": 1024,
156
+ "prefix": null,
157
+ "problem_type": null,
158
+ "pruned_heads": {},
159
+ "remove_invalid_values": false,
160
+ "repetition_penalty": 1.0,
161
+ "return_dict": true,
162
+ "return_dict_in_generate": false,
163
+ "rope_embeddings": false,
164
+ "rope_theta": 10000.0,
165
+ "scale_embedding": false,
166
+ "sep_token_id": null,
167
+ "suppress_tokens": null,
168
+ "task_specific_params": null,
169
+ "temperature": 1.0,
170
+ "tf_legacy_loss": false,
171
+ "tie_encoder_decoder": false,
172
+ "tie_word_embeddings": false,
173
+ "tokenizer_class": null,
174
+ "top_k": 50,
175
+ "top_p": 1.0,
176
+ "torch_dtype": "float32",
177
+ "torchscript": false,
178
+ "typical_p": 1.0,
179
+ "use_bfloat16": false,
180
+ "use_cache": true,
181
+ "use_fused_lm_heads": true,
182
+ "vocab_size": 1088
183
+ },
184
+ "decoder_start_token_id": 1025,
185
+ "is_encoder_decoder": true,
186
+ "model_type": "parler_tts",
187
+ "pad_token_id": 1024,
188
+ "prompt_cross_attention": false,
189
+ "text_encoder": {
190
+ "_attn_implementation_autoset": false,
191
+ "_name_or_path": "google/flan-t5-large",
192
+ "add_cross_attention": false,
193
+ "architectures": [
194
+ "T5ForConditionalGeneration"
195
+ ],
196
+ "bad_words_ids": null,
197
+ "begin_suppress_tokens": null,
198
+ "bos_token_id": null,
199
+ "chunk_size_feed_forward": 0,
200
+ "classifier_dropout": 0.0,
201
+ "cross_attention_hidden_size": null,
202
+ "d_ff": 2816,
203
+ "d_kv": 64,
204
+ "d_model": 1024,
205
+ "decoder_start_token_id": 0,
206
+ "dense_act_fn": "gelu_new",
207
+ "diversity_penalty": 0.0,
208
+ "do_sample": false,
209
+ "dropout_rate": 0.1,
210
+ "early_stopping": false,
211
+ "encoder_no_repeat_ngram_size": 0,
212
+ "eos_token_id": 1,
213
+ "exponential_decay_length_penalty": null,
214
+ "feed_forward_proj": "gated-gelu",
215
+ "finetuning_task": null,
216
+ "forced_bos_token_id": null,
217
+ "forced_eos_token_id": null,
218
+ "id2label": {
219
+ "0": "LABEL_0",
220
+ "1": "LABEL_1"
221
+ },
222
+ "initializer_factor": 1.0,
223
+ "is_decoder": false,
224
+ "is_encoder_decoder": true,
225
+ "is_gated_act": true,
226
+ "label2id": {
227
+ "LABEL_0": 0,
228
+ "LABEL_1": 1
229
+ },
230
+ "layer_norm_epsilon": 1e-06,
231
+ "length_penalty": 1.0,
232
+ "max_length": 20,
233
+ "min_length": 0,
234
+ "model_type": "t5",
235
+ "n_positions": 512,
236
+ "no_repeat_ngram_size": 0,
237
+ "num_beam_groups": 1,
238
+ "num_beams": 1,
239
+ "num_decoder_layers": 24,
240
+ "num_heads": 16,
241
+ "num_layers": 24,
242
+ "num_return_sequences": 1,
243
+ "output_attentions": false,
244
+ "output_hidden_states": false,
245
+ "output_past": true,
246
+ "output_scores": false,
247
+ "pad_token_id": 0,
248
+ "prefix": null,
249
+ "problem_type": null,
250
+ "pruned_heads": {},
251
+ "relative_attention_max_distance": 128,
252
+ "relative_attention_num_buckets": 32,
253
+ "remove_invalid_values": false,
254
+ "repetition_penalty": 1.0,
255
+ "return_dict": true,
256
+ "return_dict_in_generate": false,
257
+ "sep_token_id": null,
258
+ "suppress_tokens": null,
259
+ "task_specific_params": null,
260
+ "temperature": 1.0,
261
+ "tf_legacy_loss": false,
262
+ "tie_encoder_decoder": false,
263
+ "tie_word_embeddings": false,
264
+ "tokenizer_class": null,
265
+ "top_k": 50,
266
+ "top_p": 1.0,
267
+ "torch_dtype": null,
268
+ "torchscript": false,
269
+ "typical_p": 1.0,
270
+ "use_bfloat16": false,
271
+ "use_cache": true,
272
+ "vocab_size": 32128
273
+ },
274
+ "torch_dtype": "float16",
275
+ "transformers_version": "4.46.1",
276
+ "vocab_size": 90714
277
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
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
+ "max_length": 2610,
8
+ "pad_token_id": 1024,
9
+ "transformers_version": "4.46.1"
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bea081a5a44a6bf4520c9cb4f3c815f07b5a978826224a55e2ec98ba928c1f00
3
+ size 1875714666
preprocessor_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length_s": null,
3
+ "feature_extractor_type": "EncodecFeatureExtractor",
4
+ "feature_size": 1,
5
+ "overlap": null,
6
+ "padding_side": "right",
7
+ "padding_value": 0.0,
8
+ "return_attention_mask": true,
9
+ "sampling_rate": 44100
10
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<unk>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a356681c7eeca5f2b65c0a151da2d5c1c79f2f965517883d25e421615fe79ada
3
+ size 10971770
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bc8fa773221597d09cfadb23a2b1bd717488a0481505469ea56d42cb044de9b5
3
+ size 1795391
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": false,
35
+ "model_max_length": 1000000000000000019884624838656,
36
+ "pad_token": "<unk>",
37
+ "padding_side": "left",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }