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

phoneme punctuation bug fix

This commit is contained in:
Eren Golge 2019-02-16 03:20:04 +01:00
parent 4d2c374cfd
commit 97a16cedbf
2 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
utils/text/__init__.py
View File

@ -4,7 +4,7 @@ import re
import phonemizer import phonemizer
from phonemizer.phonemize import phonemize from phonemizer.phonemize import phonemize
from utils.text import cleaners from utils.text import cleaners
from utils.text.symbols import symbols, phonemes, _punctuations from utils.text.symbols import symbols, phonemes, _phoneme_punctuations
# Mappings from symbol to numeric ID and vice versa: # Mappings from symbol to numeric ID and vice versa:
_symbol_to_id = {s: i for i, s in enumerate(symbols)} _symbol_to_id = {s: i for i, s in enumerate(symbols)}
@ -17,7 +17,7 @@ _id_to_phonemes = {i: s for i, s in enumerate(phonemes)}
_curly_re = re.compile(r'(.*?)\{(.+?)\}(.*)') _curly_re = re.compile(r'(.*?)\{(.+?)\}(.*)')
# Regular expression matchinf punctuations, ignoring empty space # Regular expression matchinf punctuations, ignoring empty space
pat = r'['+_punctuations[:-1]+']+' pat = r'['+_phoneme_punctuations[:-1]+']+'
def text2phone(text, language): def text2phone(text, language):
@ -31,7 +31,7 @@ def text2phone(text, language):
# Replace \n with matching punctuations. # Replace \n with matching punctuations.
if len(punctuations) > 0: if len(punctuations) > 0:
for punct in punctuations[:-1]: for punct in punctuations[:-1]:
ph = ph.replace(' \n', punct+'| ', 1) ph = ph.replace('| |\n', '|'+punct+'| |', 1)
try: try:
ph = ph[:-1] + punctuations[-1] ph = ph[:-1] + punctuations[-1]
except: except:
@ -63,7 +63,6 @@ def sequence_to_phoneme(sequence):
for symbol_id in sequence: for symbol_id in sequence:
if symbol_id in _id_to_phonemes: if symbol_id in _id_to_phonemes:
s = _id_to_phonemes[symbol_id] s = _id_to_phonemes[symbol_id]
print(s)
result += s result += s
return result.replace('}{', ' ') return result.replace('}{', ' ')

1
utils/text/symbols.py
View File

@ -11,6 +11,7 @@ _pad = '_'
_eos = '~' _eos = '~'
_characters = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz!\'(),-.:;? ' _characters = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz!\'(),-.:;? '
_punctuations = '!\'(),-.:;? ' _punctuations = '!\'(),-.:;? '
_phoneme_punctuations = '.!;:,?'
# TODO: include more phoneme characters for other languages. # TODO: include more phoneme characters for other languages.
_phonemes = ['l','ɹ','ɜ','ɚ','k','u','ʔ','ð','ɐ','ɾ','ɑ','ɔ','b','ɛ','t','v','n','m','ʊ','ŋ','s', _phonemes = ['l','ɹ','ɜ','ɚ','k','u','ʔ','ð','ɐ','ɾ','ɑ','ɔ','b','ɛ','t','v','n','m','ʊ','ŋ','s',