-
Notifications
You must be signed in to change notification settings - Fork 1
/
data_utils.py
164 lines (135 loc) · 6.48 KB
/
data_utils.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
import random
import librosa
import numpy as np
import torch
import torch.utils.data
import audio
import layers
from utils import load_wav_to_torch, load_filepaths_and_text, get_split_mels, split_audio
from text import text_to_sequence
_pad = 0
class TextMelLoader(torch.utils.data.Dataset):
"""
1) loads audio,text pairs
2) normalizes text and converts them to sequences of one-hot vectors
3) computes mel-spectrograms from audio files.
"""
# pair ==>[]
def __init__(self, audiopaths_and_text, hparams, shuffle=True):
self.audiopaths_and_text = load_filepaths_and_text(
audiopaths_and_text, hparams.sort_by_length)
self.text_cleaners = hparams.text_cleaners
self.max_wav_value = hparams.max_wav_value
self.sampling_rate = hparams.sampling_rate
self.load_mel_from_disk = hparams.load_mel_from_disk
self.speaker_encoder = layers.SpeakerEncoder(hparams.num_mel, )
self.speaker_encoder.load_model(hparams.se_checkpoint)
self.speaker_encoder.eval()
self.hparms = hparams
self.stft = layers.TacotronSTFT(
hparams.filter_length, hparams.hop_length, hparams.win_length,
hparams.n_mel_channels, hparams.sampling_rate, hparams.mel_fmin,
hparams.mel_fmax)
random.seed(1234)
if shuffle:
random.shuffle(self.audiopaths_and_text)
def get_mel_text_pair(self, audiopath_and_text):
# separate filename and text
audiopath, text = audiopath_and_text[0], audiopath_and_text[1]
text = self.get_text(text)
speaker_encoder, spectrum, mel = self.get_mel(audiopath)
return (text, mel, spectrum, speaker_encoder)
def get_mel(self, filename):
if not self.load_mel_from_disk:
wav,_ = librosa.load(filename, self.sampling_rate)
wav = torch.from_numpy(wav).float().unsqueeze(0)
#audio_norm = wav / self.max_wav_value
#audio_norm = audio_norm.unsqueeze(0)
audio_norm = torch.autograd.Variable(wav, requires_grad=False)
melspec = self.stft.mel_spectrogram(audio_norm)
melspec = torch.squeeze(melspec, 0)
wav, sr = librosa.load(filename, sr=self.hparms.se_sample_rate)
wav, _ = librosa.effects.trim(wav, top_db=20)
audios = split_audio(wav, sr=self.hparms.se_sample_rate, )
mels = get_split_mels(audios,
# sr=self.hparms.se_sample_rate,
# n_fft=self.hparms.se_n_fft,
# win_length=self.hparms.se_window,
# hop_length=self.hparms.se_hop,
mel=self.hparms.num_mel)
if len(mels)==0:
print(filename)
mels = np.stack(mels)
mels = torch.from_numpy(mels).float()
mels = mels.permute(0, 2, 1)
x, _ = self.speaker_encoder(mels, return_sim=False)
speaker_encoder = x.mean(0) # final speaker encode from an audio
# reference from gst
spectrogram = audio.spectrogram(wav).astype(np.float32)
spectrogram = spectrogram.transpose(1,0)
else:
melspec = torch.from_numpy(np.load(filename))
assert melspec.size(0) == self.stft.n_mel_channels, (
'Mel dimension mismatch: given {}, expected {}'.format(
melspec.size(0), self.stft.n_mel_channels))
return speaker_encoder, spectrogram, melspec
def get_text(self, text):
text_norm = torch.IntTensor(text_to_sequence(text, self.text_cleaners))
return text_norm
def __getitem__(self, index):
return self.get_mel_text_pair(self.audiopaths_and_text[index])
def __len__(self):
return len(self.audiopaths_and_text)
class TextMelCollate():
""" Zero-pads model inputs and targets based on number of frames per setep
"""
def __init__(self, n_frames_per_step):
self.n_frames_per_step = n_frames_per_step
def __call__(self, batch):
"""Collate's training batch from normalized text and mel-spectrogram,
spectrum and 256-dim speaker_encoder
PARAMS
------
batch: [text_normalized, mel_normalized]
"""
# Right zero-pad all one-hot text sequences to max input length
input_lengths, ids_sorted_decreasing = torch.sort(
torch.LongTensor([len(x[0]) for x in batch]),
dim=0, descending=True)
max_input_len = input_lengths[0]
text_padded = torch.LongTensor(len(batch), max_input_len)
text_padded.zero_()
for i in range(len(ids_sorted_decreasing)):
text = batch[ids_sorted_decreasing[i]][0]
text_padded[i, :text.size(0)] = text
spectrum = _prepare_targets([x[2] for x in batch], self.n_frames_per_step)
d_vector = [x[3] for x in batch]
d_vector = torch.stack(d_vector)
# Right zero-pad mel-spec with extra single zero vector to mark the end
num_mels = batch[0][1].size(0)
max_target_len = max([x[1].size(1) for x in batch]) + 1
if max_target_len % self.n_frames_per_step != 0:
max_target_len += self.n_frames_per_step - max_target_len % self.n_frames_per_step
assert max_target_len % self.n_frames_per_step == 0
# include mel padded and gate padded
mel_padded = torch.FloatTensor(len(batch), num_mels, max_target_len)
mel_padded.zero_()
gate_padded = torch.FloatTensor(len(batch), max_target_len)
gate_padded.zero_()
output_lengths = torch.LongTensor(len(batch))
for i in range(len(ids_sorted_decreasing)):
mel = batch[ids_sorted_decreasing[i]][1]
mel_padded[i, :, :mel.size(1)] = mel
gate_padded[i, mel.size(1):] = 1
output_lengths[i] = mel.size(1)
return text_padded, input_lengths, mel_padded, gate_padded, output_lengths, spectrum, d_vector
def _prepare_targets(targets, alignment):
max_len = max((t.shape[0] for t in targets)) + 1
return np.stack([_pad_target(t, _round_up(max_len, alignment)) for t in targets])
def _pad_input(x, length):
return np.pad(x, (0, length - x.shape[0]), mode='constant', constant_values=_pad)
def _pad_target(t, length):
return np.pad(t, [(0, length - t.shape[0]), (0, 0)], mode='constant', constant_values=_pad)
def _round_up(x, multiple):
remainder = x % multiple
return x if remainder == 0 else x + multiple - remainder