phonem extraction for training

datasets/TTSDataset.py

@@ -87,7 +87,7 @@ class MyDataset(Dataset):
         else:
             text, wav_file = self.items[idx]
             text = np.asarray(
-                text_to_sequence(text, [self.cleaners]), dtype=np.int32)
+                phoneme_to_sequence(text, [self.cleaners]), dtype=np.int32)
             wav = np.asarray(self.load_wav(wav_file), dtype=np.float32)
             sample = {'text': text, 'wav': wav, 'item_idx': self.items[idx][1]}
         return sample

utils/text/__init__.py

@@ -3,7 +3,11 @@
 import re
 import phonemizer
 from utils.text import cleaners
+<<<<<<< HEAD
 from utils.text.symbols import symbols, phonemes, _punctuations
+=======
+from utils.text.symbols import symbols, phonemes
+>>>>>>> phonem extraction for training
 
 # Mappings from symbol to numeric ID and vice versa:
 _symbol_to_id = {s: i for i, s in enumerate(symbols)}
@@ -67,6 +71,13 @@ def sequence_to_phoneme(sequence):
     return result.replace('}{', ' ')
 
 
+def phoneme_to_sequence(text, cleaner_names):
+    sequence = []
+    sequence += _phonem_to_sequence(_clean_text(text, cleaner_names))
+    sequence.append(_phonemes_to_id['~'])
+    return sequence
+
+
 def text_to_sequence(text, cleaner_names):
     '''Converts a string of text to a sequence of IDs corresponding to the symbols in the text.
 

utils/text/cmudict.py

@@ -1,6 +1,7 @@
 # -*- coding: utf-8 -*-
 
 import re
+import phonemizer
 
 # valid_symbols = [
 #     'AA', 'AA0', 'AA1', 'AA2', 'AE', 'AE0', 'AE1', 'AE2', 'AH', 'AH0', 'AH1',