-
Notifications
You must be signed in to change notification settings - Fork 4
/
asv.py
191 lines (157 loc) · 5.35 KB
/
asv.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
# Copyright 3D-Speaker (https://github.com/alibaba-damo-academy/3D-Speaker). All Rights Reserved.
# Licensed under the Apache License, Version 2.0 (http://www.apache.org/licenses/LICENSE-2.0)
"""
This script will download pretrained models from modelscope (https://www.modelscope.cn/models)
based on the given model id, and extract embeddings from input audio.
Please pre-install "modelscope".
Usage:
1. extract the embedding from the wav file.
`python infer_sv.py --model_id $model_id --wavs $wav_path `
2. extract embeddings from two wav files and compute the similarity score.
`python infer_sv.py --model_id $model_id --wavs $wav_path1 $wav_path2 `
3. extract embeddings from the wav list.
`python infer_sv.py --model_id $model_id --wavs $wav_list `
"""
import os
import sys
import re
import pathlib
import numpy as np
import argparse
import torch
import torchaudio
try:
from speakerlab.process.processor import FBank
except ImportError:
sys.path.append('%s/../..'%os.path.dirname(__file__))
from speakerlab.process.processor import FBank
from speakerlab.utils.builder import dynamic_import
from modelscope.hub.snapshot_download import snapshot_download
from modelscope.pipelines.util import is_official_hub_path
CAMPPLUS_VOX = {
'obj': 'speakerlab.models.campplus.DTDNN.CAMPPlus',
'args': {
'feat_dim': 80,
'embedding_size': 512,
},
}
CAMPPLUS_COMMON = {
'obj': 'speakerlab.models.campplus.DTDNN.CAMPPlus',
'args': {
'feat_dim': 80,
'embedding_size': 192,
},
}
ERes2Net_VOX = {
'obj': 'speakerlab.models.eres2net.ResNet.ERes2Net',
'args': {
'feat_dim': 80,
'embedding_size': 192,
},
}
ERes2Net_COMMON = {
'obj': 'speakerlab.models.eres2net.ResNet_aug.ERes2Net',
'args': {
'feat_dim': 80,
'embedding_size': 192,
},
}
ERes2Net_base_COMMON = {
'obj': 'speakerlab.models.eres2net.ResNet.ERes2Net',
'args': {
'feat_dim': 80,
'embedding_size': 512,
'm_channels': 32,
},
}
ERes2Net_Base_3D_Speaker = {
'obj': 'speakerlab.models.eres2net.ResNet.ERes2Net',
'args': {
'feat_dim': 80,
'embedding_size': 512,
'm_channels': 32,
},
}
ERes2Net_Large_3D_Speaker = {
'obj': 'speakerlab.models.eres2net.ResNet.ERes2Net',
'args': {
'feat_dim': 80,
'embedding_size': 512,
'm_channels': 64,
},
}
supports = {
'damo/speech_campplus_sv_en_voxceleb_16k': {
'revision': 'v1.0.2',
'model': CAMPPLUS_VOX,
'model_pt': 'campplus_voxceleb.bin',
},
'damo/speech_campplus_sv_zh-cn_16k-common': {
'revision': 'v1.0.0',
'model': CAMPPLUS_COMMON,
'model_pt': 'campplus_cn_common.bin',
},
'damo/speech_eres2net_sv_en_voxceleb_16k': {
'revision': 'v1.0.2',
'model': ERes2Net_VOX,
'model_pt': 'pretrained_eres2net.ckpt',
},
'damo/speech_eres2net_sv_zh-cn_16k-common': {
'revision': 'v1.0.4',
'model': ERes2Net_COMMON,
'model_pt': 'pretrained_eres2net_aug.ckpt',
},
'damo/speech_eres2net_base_200k_sv_zh-cn_16k-common': {
'revision': 'v1.0.0',
'model': ERes2Net_base_COMMON,
'model_pt': 'pretrained_eres2net.pt',
},
'damo/speech_eres2net_base_sv_zh-cn_3dspeaker_16k': {
'revision': 'v1.0.1',
'model': ERes2Net_Base_3D_Speaker,
'model_pt': 'eres2net_base_model.ckpt',
},
'damo/speech_eres2net_large_sv_zh-cn_3dspeaker_16k': {
'revision': 'v1.0.0',
'model': ERes2Net_Large_3D_Speaker,
'model_pt': 'eres2net_large_model.ckpt',
},
}
def get_model(model_id, local_model_dir):
conf = supports[model_id]
save_dir = os.path.join(local_model_dir, model_id.split('/')[1])
save_dir = pathlib.Path(save_dir)
pretrained_model = save_dir / conf['model_pt']
pretrained_state = torch.load(pretrained_model, map_location='cpu')
# load model
model = conf['model']
embedding_model = dynamic_import(model['obj'])(**model['args'])
embedding_model.load_state_dict(pretrained_state)
embedding_model.eval()
return embedding_model
def get_asv_models(model_ids, local_model_dir):
models = []
for model_id in model_ids:
models.append(get_model(model_id, local_model_dir))
feature_extractor = FBank(80, sample_rate=16000, mean_nor=True)
return models, feature_extractor
def compute_embedding(wav, embedding_model, feature_extractor):
# compute feat
feat = feature_extractor(wav).unsqueeze(0)
# compute embedding
with torch.no_grad():
embedding = embedding_model(feat)
return embedding
def compute_similarity(wav1, wav2, embedding_model, feature_extractor):
embedding1 = compute_embedding(wav1, embedding_model, feature_extractor)
embedding2 = compute_embedding(wav2, embedding_model, feature_extractor)
# compute similarity score
similarity = torch.nn.CosineSimilarity(dim=-1, eps=1e-6)
scores = similarity(embedding1, embedding2).item()
return scores
def compute_similarity2(wav1, embedding2, embedding_model, feature_extractor):
embedding1 = compute_embedding(wav1, embedding_model, feature_extractor)
# compute similarity score
similarity = torch.nn.CosineSimilarity(dim=-1, eps=1e-6)
scores = similarity(embedding1, embedding2).item()
return scores