mrq/DL-Art-School
1
3
Fork 5
Code Issues Pull Requests Packages Projects Releases Wiki Activity

can i has fix now

Browse Source
This commit is contained in:
James Betker 2022-01-01 00:48:27 -07:00
parent 56752f1dbc
commit 4d9ba4a48a
1 changed files with 5 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
codes/data/audio/grand_conjoined_dataset.py
View File

@ -28,10 +28,6 @@ def build_paired_voice_dataset(args):
return D(dataset_opt) return D(dataset_opt)
def clamp(x, minimum, maximum):
return max(minimum, min(x, maximum))
class GrandConjoinedDataset(torch.utils.data.Dataset): class GrandConjoinedDataset(torch.utils.data.Dataset):
""" """
A joint text & speech dataset that joins three separate datasets into a single batch: A joint text & speech dataset that joins three separate datasets into a single batch:
@ -78,7 +74,7 @@ class GrandConjoinedDataset(torch.utils.data.Dataset):
unsupervised_audio_args['resample_clip'] = False unsupervised_audio_args['resample_clip'] = False
unsupervised_audio_args['extra_samples'] = self.num_conditioning_candidates unsupervised_audio_args['extra_samples'] = self.num_conditioning_candidates
unsupervised_audio_args['extra_sample_length'] = self.conditioning_length unsupervised_audio_args['extra_sample_length'] = self.conditioning_length
if self.collate: if not self.collate:
unsupervised_audio_args['pad_to_samples'] = self.max_solo_audio_length unsupervised_audio_args['pad_to_samples'] = self.max_solo_audio_length
self.speech = UnsupervisedAudioDataset(unsupervised_audio_args) self.speech = UnsupervisedAudioDataset(unsupervised_audio_args)
self.text = HfDataset(**text_corpus_args) self.text = HfDataset(**text_corpus_args)
@ -148,7 +144,7 @@ class GrandConjoinedDataset(torch.utils.data.Dataset):
sp = self.speech[i % len(self.speech)] sp = self.speech[i % len(self.speech)]
# Set upper bound on solo speech lengths. This is handled automatically when collation is turned off, but needs to be done otherwise. # Set upper bound on solo speech lengths. This is handled automatically when collation is turned off, but needs to be done otherwise.
sp['clip'] = sp['clip'][:, :self.max_solo_audio_length] sp['clip'] = sp['clip'][:, :self.max_solo_audio_length]
sp['clip_lengths'] = clamp(sp['clip_lengths'], 0, self.max_solo_audio_length) sp['clip_lengths'] = sp['clip_lengths'].clamp(0, self.max_solo_audio_length)
return self.optionally_add_conditioning_candidates({ return self.optionally_add_conditioning_candidates({
'paired_audio': snt['wav'], 'paired_audio': snt['wav'],
'paired_audio_lengths': snt['wav_lengths'], 'paired_audio_lengths': snt['wav_lengths'],
@ -181,8 +177,8 @@ if __name__ == '__main__':
'max_paired_text_length': 200, 'max_paired_text_length': 200,
'max_solo_text_length': 330, 'max_solo_text_length': 330,
'max_solo_audio_length': 300000, 'max_solo_audio_length': 300000,
'needs_collate': True, 'needs_collate': False,
'num_conditioning_candidates': 2, 'num_conditioning_candidates': 1,
'conditioning_length': 44000, 'conditioning_length': 44000,
'paired_dataset_args': { 'paired_dataset_args': {
'path': ['Y:\\clips\\podcasts-0-transcribed.tsv'], 'path': ['Y:\\clips\\podcasts-0-transcribed.tsv'],
@ -209,7 +205,7 @@ if __name__ == '__main__':
'max_solo_text_length': 330, 'max_solo_text_length': 330,
'max_solo_audio_length': 300000, 'max_solo_audio_length': 300000,
'only_paired': True, 'only_paired': True,
'needs_collate': True, 'needs_collate': False,
'paired_dataset_args': { 'paired_dataset_args': {
'path': ['Z:\\bigasr_dataset\\libritts\\test-clean_list.txt'], 'path': ['Z:\\bigasr_dataset\\libritts\\test-clean_list.txt'],
'fetcher_mode': ['libritts'], 'fetcher_mode': ['libritts'],