Commit
•
d576da5
1
Parent(s):
ac620ec
Add model weights and configs
Browse files- feature_extractor/preprocessor_config.json +22 -0
- model_index.json +32 -0
- scheduler/scheduler_config.json +19 -0
- text_encoder/config.json +206 -0
- text_encoder/pytorch_model.bin +3 -0
- tokenizer/merges.txt +0 -0
- tokenizer/preprocessor_config.json +22 -0
- tokenizer/special_tokens_map.json +15 -0
- tokenizer/tokenizer.json +0 -0
- tokenizer/tokenizer_config.json +16 -0
- tokenizer/vocab.json +0 -0
- unet/config.json +69 -0
- unet/diffusion_pytorch_model.bin +3 -0
- vae/config.json +28 -0
- vae/diffusion_pytorch_model.bin +3 -0
- vocoder/config.json +50 -0
- vocoder/pytorch_model.bin +3 -0
feature_extractor/preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chunk_length_s": 10,
|
3 |
+
"feature_extractor_type": "ClapFeatureExtractor",
|
4 |
+
"feature_size": 64,
|
5 |
+
"fft_window_size": 1024,
|
6 |
+
"frequency_max": 14000,
|
7 |
+
"frequency_min": 50,
|
8 |
+
"hop_length": 480,
|
9 |
+
"max_length_s": 10,
|
10 |
+
"n_fft": 1024,
|
11 |
+
"nb_frequency_bins": 513,
|
12 |
+
"nb_max_frames": 1000,
|
13 |
+
"nb_max_samples": 480000,
|
14 |
+
"padding": "repeatpad",
|
15 |
+
"padding_side": "right",
|
16 |
+
"padding_value": 0.0,
|
17 |
+
"processor_class": "ClapProcessor",
|
18 |
+
"return_attention_mask": false,
|
19 |
+
"sampling_rate": 48000,
|
20 |
+
"top_db": null,
|
21 |
+
"truncation": "rand_trunc"
|
22 |
+
}
|
model_index.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_class_name": "MusicLDMPipeline",
|
3 |
+
"_diffusers_version": "0.20.0.dev0",
|
4 |
+
"feature_extractor": [
|
5 |
+
"transformers",
|
6 |
+
"ClapFeatureExtractor"
|
7 |
+
],
|
8 |
+
"scheduler": [
|
9 |
+
"diffusers",
|
10 |
+
"DDIMScheduler"
|
11 |
+
],
|
12 |
+
"text_encoder": [
|
13 |
+
"transformers",
|
14 |
+
"ClapModel"
|
15 |
+
],
|
16 |
+
"tokenizer": [
|
17 |
+
"transformers",
|
18 |
+
"RobertaTokenizerFast"
|
19 |
+
],
|
20 |
+
"unet": [
|
21 |
+
"diffusers",
|
22 |
+
"UNet2DConditionModel"
|
23 |
+
],
|
24 |
+
"vae": [
|
25 |
+
"diffusers",
|
26 |
+
"AutoencoderKL"
|
27 |
+
],
|
28 |
+
"vocoder": [
|
29 |
+
"transformers",
|
30 |
+
"SpeechT5HifiGan"
|
31 |
+
]
|
32 |
+
}
|
scheduler/scheduler_config.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_class_name": "DDIMScheduler",
|
3 |
+
"_diffusers_version": "0.20.0.dev0",
|
4 |
+
"beta_end": 0.0195,
|
5 |
+
"beta_schedule": "scaled_linear",
|
6 |
+
"beta_start": 0.0015,
|
7 |
+
"clip_sample": false,
|
8 |
+
"clip_sample_range": 1.0,
|
9 |
+
"dynamic_thresholding_ratio": 0.995,
|
10 |
+
"num_train_timesteps": 1000,
|
11 |
+
"prediction_type": "epsilon",
|
12 |
+
"rescale_betas_zero_snr": false,
|
13 |
+
"sample_max_value": 1.0,
|
14 |
+
"set_alpha_to_one": false,
|
15 |
+
"steps_offset": 1,
|
16 |
+
"thresholding": false,
|
17 |
+
"timestep_spacing": "leading",
|
18 |
+
"trained_betas": null
|
19 |
+
}
|
text_encoder/config.json
ADDED
@@ -0,0 +1,206 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_commit_hash": "8fa0f1c6d0433df6e97c127f64b2a1d6c0dcda8a",
|
3 |
+
"architectures": [
|
4 |
+
"ClapModel"
|
5 |
+
],
|
6 |
+
"audio_config": {
|
7 |
+
"_name_or_path": "",
|
8 |
+
"add_cross_attention": false,
|
9 |
+
"aff_block_r": 4,
|
10 |
+
"architectures": null,
|
11 |
+
"attention_probs_dropout_prob": 0.0,
|
12 |
+
"bad_words_ids": null,
|
13 |
+
"begin_suppress_tokens": null,
|
14 |
+
"bos_token_id": null,
|
15 |
+
"chunk_size_feed_forward": 0,
|
16 |
+
"cross_attention_hidden_size": null,
|
17 |
+
"decoder_start_token_id": null,
|
18 |
+
"depths": [
|
19 |
+
2,
|
20 |
+
2,
|
21 |
+
12,
|
22 |
+
2
|
23 |
+
],
|
24 |
+
"diversity_penalty": 0.0,
|
25 |
+
"do_sample": false,
|
26 |
+
"drop_path_rate": 0.0,
|
27 |
+
"early_stopping": false,
|
28 |
+
"enable_fusion": false,
|
29 |
+
"enable_patch_layer_norm": true,
|
30 |
+
"encoder_no_repeat_ngram_size": 0,
|
31 |
+
"eos_token_id": null,
|
32 |
+
"exponential_decay_length_penalty": null,
|
33 |
+
"finetuning_task": null,
|
34 |
+
"flatten_patch_embeds": true,
|
35 |
+
"forced_bos_token_id": null,
|
36 |
+
"forced_eos_token_id": null,
|
37 |
+
"fusion_num_hidden_layers": 2,
|
38 |
+
"fusion_type": null,
|
39 |
+
"hidden_act": "gelu",
|
40 |
+
"hidden_dropout_prob": 0.1,
|
41 |
+
"hidden_size": 1024,
|
42 |
+
"id2label": {
|
43 |
+
"0": "LABEL_0",
|
44 |
+
"1": "LABEL_1"
|
45 |
+
},
|
46 |
+
"initializer_factor": 1.0,
|
47 |
+
"is_decoder": false,
|
48 |
+
"is_encoder_decoder": false,
|
49 |
+
"label2id": {
|
50 |
+
"LABEL_0": 0,
|
51 |
+
"LABEL_1": 1
|
52 |
+
},
|
53 |
+
"layer_norm_eps": 1e-05,
|
54 |
+
"length_penalty": 1.0,
|
55 |
+
"max_length": 20,
|
56 |
+
"min_length": 0,
|
57 |
+
"mlp_ratio": 4.0,
|
58 |
+
"model_type": "clap_audio_model",
|
59 |
+
"no_repeat_ngram_size": 0,
|
60 |
+
"num_attention_heads": [
|
61 |
+
4,
|
62 |
+
8,
|
63 |
+
16,
|
64 |
+
32
|
65 |
+
],
|
66 |
+
"num_beam_groups": 1,
|
67 |
+
"num_beams": 1,
|
68 |
+
"num_classes": 527,
|
69 |
+
"num_hidden_layers": 4,
|
70 |
+
"num_mel_bins": 64,
|
71 |
+
"num_return_sequences": 1,
|
72 |
+
"output_attentions": false,
|
73 |
+
"output_hidden_states": false,
|
74 |
+
"output_scores": false,
|
75 |
+
"pad_token_id": null,
|
76 |
+
"patch_embed_input_channels": 1,
|
77 |
+
"patch_embeds_hidden_size": 128,
|
78 |
+
"patch_size": 4,
|
79 |
+
"patch_stride": [
|
80 |
+
4,
|
81 |
+
4
|
82 |
+
],
|
83 |
+
"prefix": null,
|
84 |
+
"problem_type": null,
|
85 |
+
"projection_dim": 512,
|
86 |
+
"projection_hidden_act": "relu",
|
87 |
+
"projection_hidden_size": 768,
|
88 |
+
"pruned_heads": {},
|
89 |
+
"qkv_bias": true,
|
90 |
+
"remove_invalid_values": false,
|
91 |
+
"repetition_penalty": 1.0,
|
92 |
+
"return_dict": true,
|
93 |
+
"return_dict_in_generate": false,
|
94 |
+
"sep_token_id": null,
|
95 |
+
"spec_size": 256,
|
96 |
+
"suppress_tokens": null,
|
97 |
+
"task_specific_params": null,
|
98 |
+
"temperature": 1.0,
|
99 |
+
"tf_legacy_loss": false,
|
100 |
+
"tie_encoder_decoder": false,
|
101 |
+
"tie_word_embeddings": true,
|
102 |
+
"tokenizer_class": null,
|
103 |
+
"top_k": 50,
|
104 |
+
"top_p": 1.0,
|
105 |
+
"torch_dtype": null,
|
106 |
+
"torchscript": false,
|
107 |
+
"transformers_version": "4.30.2",
|
108 |
+
"typical_p": 1.0,
|
109 |
+
"use_bfloat16": false,
|
110 |
+
"window_size": 8
|
111 |
+
},
|
112 |
+
"hidden_size": 768,
|
113 |
+
"initializer_factor": 1.0,
|
114 |
+
"logit_scale_init_value": 14.285714285714285,
|
115 |
+
"model_type": "clap",
|
116 |
+
"num_hidden_layers": 16,
|
117 |
+
"projection_dim": 512,
|
118 |
+
"projection_hidden_act": "relu",
|
119 |
+
"text_config": {
|
120 |
+
"_name_or_path": "",
|
121 |
+
"add_cross_attention": false,
|
122 |
+
"architectures": null,
|
123 |
+
"attention_probs_dropout_prob": 0.1,
|
124 |
+
"bad_words_ids": null,
|
125 |
+
"begin_suppress_tokens": null,
|
126 |
+
"bos_token_id": 0,
|
127 |
+
"chunk_size_feed_forward": 0,
|
128 |
+
"classifier_dropout": null,
|
129 |
+
"cross_attention_hidden_size": null,
|
130 |
+
"decoder_start_token_id": null,
|
131 |
+
"diversity_penalty": 0.0,
|
132 |
+
"do_sample": false,
|
133 |
+
"early_stopping": false,
|
134 |
+
"encoder_no_repeat_ngram_size": 0,
|
135 |
+
"eos_token_id": 2,
|
136 |
+
"exponential_decay_length_penalty": null,
|
137 |
+
"finetuning_task": null,
|
138 |
+
"forced_bos_token_id": null,
|
139 |
+
"forced_eos_token_id": null,
|
140 |
+
"fusion_hidden_size": 768,
|
141 |
+
"fusion_num_hidden_layers": 2,
|
142 |
+
"hidden_act": "gelu",
|
143 |
+
"hidden_dropout_prob": 0.1,
|
144 |
+
"hidden_size": 768,
|
145 |
+
"id2label": {
|
146 |
+
"0": "LABEL_0",
|
147 |
+
"1": "LABEL_1"
|
148 |
+
},
|
149 |
+
"initializer_factor": 1.0,
|
150 |
+
"initializer_range": 0.02,
|
151 |
+
"intermediate_size": 3072,
|
152 |
+
"is_decoder": false,
|
153 |
+
"is_encoder_decoder": false,
|
154 |
+
"label2id": {
|
155 |
+
"LABEL_0": 0,
|
156 |
+
"LABEL_1": 1
|
157 |
+
},
|
158 |
+
"layer_norm_eps": 1e-12,
|
159 |
+
"length_penalty": 1.0,
|
160 |
+
"max_length": 20,
|
161 |
+
"max_position_embeddings": 514,
|
162 |
+
"min_length": 0,
|
163 |
+
"model_type": "clap_text_model",
|
164 |
+
"no_repeat_ngram_size": 0,
|
165 |
+
"num_attention_heads": 12,
|
166 |
+
"num_beam_groups": 1,
|
167 |
+
"num_beams": 1,
|
168 |
+
"num_hidden_layers": 12,
|
169 |
+
"num_return_sequences": 1,
|
170 |
+
"output_attentions": false,
|
171 |
+
"output_hidden_states": false,
|
172 |
+
"output_scores": false,
|
173 |
+
"pad_token_id": 1,
|
174 |
+
"position_embedding_type": "absolute",
|
175 |
+
"prefix": null,
|
176 |
+
"problem_type": null,
|
177 |
+
"projection_dim": 512,
|
178 |
+
"projection_hidden_act": "relu",
|
179 |
+
"projection_hidden_size": 768,
|
180 |
+
"pruned_heads": {},
|
181 |
+
"remove_invalid_values": false,
|
182 |
+
"repetition_penalty": 1.0,
|
183 |
+
"return_dict": true,
|
184 |
+
"return_dict_in_generate": false,
|
185 |
+
"sep_token_id": null,
|
186 |
+
"suppress_tokens": null,
|
187 |
+
"task_specific_params": null,
|
188 |
+
"temperature": 1.0,
|
189 |
+
"tf_legacy_loss": false,
|
190 |
+
"tie_encoder_decoder": false,
|
191 |
+
"tie_word_embeddings": true,
|
192 |
+
"tokenizer_class": null,
|
193 |
+
"top_k": 50,
|
194 |
+
"top_p": 1.0,
|
195 |
+
"torch_dtype": null,
|
196 |
+
"torchscript": false,
|
197 |
+
"transformers_version": "4.30.2",
|
198 |
+
"type_vocab_size": 1,
|
199 |
+
"typical_p": 1.0,
|
200 |
+
"use_bfloat16": false,
|
201 |
+
"use_cache": true,
|
202 |
+
"vocab_size": 50265
|
203 |
+
},
|
204 |
+
"torch_dtype": "float32",
|
205 |
+
"transformers_version": null
|
206 |
+
}
|
text_encoder/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f93b3b3fc745788d05b459fd60526de0a5148a2048c9e2e55dfccaf2ee70bb95
|
3 |
+
size 776444665
|
tokenizer/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer/preprocessor_config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"chunk_length_s": 10,
|
3 |
+
"feature_extractor_type": "ClapFeatureExtractor",
|
4 |
+
"feature_size": 64,
|
5 |
+
"fft_window_size": 1024,
|
6 |
+
"frequency_max": 14000,
|
7 |
+
"frequency_min": 50,
|
8 |
+
"hop_length": 480,
|
9 |
+
"max_length_s": 10,
|
10 |
+
"n_fft": 1024,
|
11 |
+
"nb_frequency_bins": 513,
|
12 |
+
"nb_max_frames": 1000,
|
13 |
+
"nb_max_samples": 480000,
|
14 |
+
"padding": "repeatpad",
|
15 |
+
"padding_side": "right",
|
16 |
+
"padding_value": 0.0,
|
17 |
+
"processor_class": "ClapProcessor",
|
18 |
+
"return_attention_mask": false,
|
19 |
+
"sampling_rate": 48000,
|
20 |
+
"top_db": null,
|
21 |
+
"truncation": "rand_trunc"
|
22 |
+
}
|
tokenizer/special_tokens_map.json
ADDED
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": "<s>",
|
3 |
+
"cls_token": "<s>",
|
4 |
+
"eos_token": "</s>",
|
5 |
+
"mask_token": {
|
6 |
+
"content": "<mask>",
|
7 |
+
"lstrip": true,
|
8 |
+
"normalized": false,
|
9 |
+
"rstrip": false,
|
10 |
+
"single_word": false
|
11 |
+
},
|
12 |
+
"pad_token": "<pad>",
|
13 |
+
"sep_token": "</s>",
|
14 |
+
"unk_token": "<unk>"
|
15 |
+
}
|
tokenizer/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer/tokenizer_config.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"bos_token": "<s>",
|
4 |
+
"clean_up_tokenization_spaces": true,
|
5 |
+
"cls_token": "<s>",
|
6 |
+
"eos_token": "</s>",
|
7 |
+
"errors": "replace",
|
8 |
+
"mask_token": "<mask>",
|
9 |
+
"model_max_length": 512,
|
10 |
+
"pad_token": "<pad>",
|
11 |
+
"processor_class": "ClapProcessor",
|
12 |
+
"sep_token": "</s>",
|
13 |
+
"tokenizer_class": "RobertaTokenizer",
|
14 |
+
"trim_offsets": true,
|
15 |
+
"unk_token": "<unk>"
|
16 |
+
}
|
tokenizer/vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
unet/config.json
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_class_name": "UNet2DConditionModel",
|
3 |
+
"_diffusers_version": "0.20.0.dev0",
|
4 |
+
"act_fn": "silu",
|
5 |
+
"addition_embed_type": null,
|
6 |
+
"addition_embed_type_num_heads": 64,
|
7 |
+
"addition_time_embed_dim": null,
|
8 |
+
"attention_head_dim": 8,
|
9 |
+
"block_out_channels": [
|
10 |
+
128,
|
11 |
+
256,
|
12 |
+
384,
|
13 |
+
640
|
14 |
+
],
|
15 |
+
"center_input_sample": false,
|
16 |
+
"class_embed_type": "simple_projection",
|
17 |
+
"class_embeddings_concat": true,
|
18 |
+
"conv_in_kernel": 3,
|
19 |
+
"conv_out_kernel": 3,
|
20 |
+
"cross_attention_dim": [
|
21 |
+
128,
|
22 |
+
256,
|
23 |
+
384,
|
24 |
+
640
|
25 |
+
],
|
26 |
+
"cross_attention_norm": null,
|
27 |
+
"down_block_types": [
|
28 |
+
"DownBlock2D",
|
29 |
+
"CrossAttnDownBlock2D",
|
30 |
+
"CrossAttnDownBlock2D",
|
31 |
+
"CrossAttnDownBlock2D"
|
32 |
+
],
|
33 |
+
"downsample_padding": 1,
|
34 |
+
"dual_cross_attention": false,
|
35 |
+
"encoder_hid_dim": null,
|
36 |
+
"encoder_hid_dim_type": null,
|
37 |
+
"flip_sin_to_cos": true,
|
38 |
+
"freq_shift": 0,
|
39 |
+
"in_channels": 8,
|
40 |
+
"layers_per_block": 2,
|
41 |
+
"mid_block_only_cross_attention": null,
|
42 |
+
"mid_block_scale_factor": 1,
|
43 |
+
"mid_block_type": "UNetMidBlock2DCrossAttn",
|
44 |
+
"norm_eps": 1e-05,
|
45 |
+
"norm_num_groups": 32,
|
46 |
+
"num_attention_heads": null,
|
47 |
+
"num_class_embeds": null,
|
48 |
+
"only_cross_attention": false,
|
49 |
+
"out_channels": 8,
|
50 |
+
"projection_class_embeddings_input_dim": 512,
|
51 |
+
"resnet_out_scale_factor": 1.0,
|
52 |
+
"resnet_skip_time_act": false,
|
53 |
+
"resnet_time_scale_shift": "default",
|
54 |
+
"sample_size": 128,
|
55 |
+
"time_cond_proj_dim": null,
|
56 |
+
"time_embedding_act_fn": null,
|
57 |
+
"time_embedding_dim": null,
|
58 |
+
"time_embedding_type": "positional",
|
59 |
+
"timestep_post_act": null,
|
60 |
+
"transformer_layers_per_block": 1,
|
61 |
+
"up_block_types": [
|
62 |
+
"CrossAttnUpBlock2D",
|
63 |
+
"CrossAttnUpBlock2D",
|
64 |
+
"CrossAttnUpBlock2D",
|
65 |
+
"UpBlock2D"
|
66 |
+
],
|
67 |
+
"upcast_attention": false,
|
68 |
+
"use_linear_projection": false
|
69 |
+
}
|
unet/diffusion_pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:298a06ae4dacc4f204edf80fe4a3eaec802b15bc24780536f6f8cfa675ef7614
|
3 |
+
size 740429981
|
vae/config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_class_name": "AutoencoderKL",
|
3 |
+
"_diffusers_version": "0.20.0.dev0",
|
4 |
+
"act_fn": "silu",
|
5 |
+
"block_out_channels": [
|
6 |
+
128,
|
7 |
+
256,
|
8 |
+
512
|
9 |
+
],
|
10 |
+
"down_block_types": [
|
11 |
+
"DownEncoderBlock2D",
|
12 |
+
"DownEncoderBlock2D",
|
13 |
+
"DownEncoderBlock2D"
|
14 |
+
],
|
15 |
+
"force_upcast": true,
|
16 |
+
"in_channels": 1,
|
17 |
+
"latent_channels": 8,
|
18 |
+
"layers_per_block": 2,
|
19 |
+
"norm_num_groups": 32,
|
20 |
+
"out_channels": 1,
|
21 |
+
"sample_size": 512,
|
22 |
+
"scaling_factor": 1.075324296951294,
|
23 |
+
"up_block_types": [
|
24 |
+
"UpDecoderBlock2D",
|
25 |
+
"UpDecoderBlock2D",
|
26 |
+
"UpDecoderBlock2D"
|
27 |
+
]
|
28 |
+
}
|
vae/diffusion_pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4ecc34b08d65bc63becf9738fa3c2ee0c3837c6aa42128ff274004d705eb02d6
|
3 |
+
size 221586761
|
vocoder/config.json
ADDED
@@ -0,0 +1,50 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"SpeechT5HifiGan"
|
4 |
+
],
|
5 |
+
"initializer_range": 0.01,
|
6 |
+
"leaky_relu_slope": 0.1,
|
7 |
+
"model_in_dim": 64,
|
8 |
+
"model_type": "hifigan",
|
9 |
+
"normalize_before": false,
|
10 |
+
"resblock_dilation_sizes": [
|
11 |
+
[
|
12 |
+
1,
|
13 |
+
3,
|
14 |
+
5
|
15 |
+
],
|
16 |
+
[
|
17 |
+
1,
|
18 |
+
3,
|
19 |
+
5
|
20 |
+
],
|
21 |
+
[
|
22 |
+
1,
|
23 |
+
3,
|
24 |
+
5
|
25 |
+
]
|
26 |
+
],
|
27 |
+
"resblock_kernel_sizes": [
|
28 |
+
3,
|
29 |
+
7,
|
30 |
+
11
|
31 |
+
],
|
32 |
+
"sampling_rate": 16000,
|
33 |
+
"torch_dtype": "float32",
|
34 |
+
"transformers_version": "4.30.2",
|
35 |
+
"upsample_initial_channel": 1024,
|
36 |
+
"upsample_kernel_sizes": [
|
37 |
+
16,
|
38 |
+
16,
|
39 |
+
8,
|
40 |
+
4,
|
41 |
+
4
|
42 |
+
],
|
43 |
+
"upsample_rates": [
|
44 |
+
5,
|
45 |
+
4,
|
46 |
+
2,
|
47 |
+
2,
|
48 |
+
2
|
49 |
+
]
|
50 |
+
}
|
vocoder/pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c1f835cc6409503d443eb01a461afa653f1ddf2703c5b43aeb74d64894a2a23d
|
3 |
+
size 221120349
|