coqui-ai
/
TTS
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Cache phoneme transcription in a tmp folder

pull/10/head
Eren Golge 2019-01-01 20:08:53 +01:00
parent 1722b1659a
commit 08de7a5b17
1 changed files with 9 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
datasets/TTSDataset.py
View File

@ -6,7 +6,7 @@ import torch
import random import random
from torch.utils.data import Dataset from torch.utils.data import Dataset
from utils.text import text_to_sequence, phonem_to_sequence from utils.text import text_to_sequence, phoneme_to_sequence
from utils.data import (prepare_data, pad_per_step, prepare_tensor, from utils.data import (prepare_data, pad_per_step, prepare_tensor,
prepare_stop_target) prepare_stop_target)
@ -86,8 +86,14 @@ class MyDataset(Dataset):
sample = {'text': text, 'wav': wav, 'item_idx': self.items[idx][1], 'mel':mel, 'linear': linear} sample = {'text': text, 'wav': wav, 'item_idx': self.items[idx][1], 'mel':mel, 'linear': linear}
else: else:
text, wav_file = self.items[idx] text, wav_file = self.items[idx]
file_name = os.path.basename(wav_file).split('.')[0]
tmp_path = os.path.join("tmp/",file_name+'_phoneme.npy')
if os.path.isfile(tmp_path):
text = np.load(tmp_path)
else:
text = np.asarray( text = np.asarray(
phonem_to_sequence(text, [self.cleaners]), dtype=np.int32) phoneme_to_sequence(text, [self.cleaners]), dtype=np.int32)
np.save(tmp_path, text)
wav = np.asarray(self.load_wav(wav_file), dtype=np.float32) wav = np.asarray(self.load_wav(wav_file), dtype=np.float32)
sample = {'text': text, 'wav': wav, 'item_idx': self.items[idx][1]} sample = {'text': text, 'wav': wav, 'item_idx': self.items[idx][1]}
return sample return sample