Публічний доступ
1
0

Improve Silero's Preprocessor to Handle Numbers and Abbreviations Better

Цей коміт міститься в:
da3dsoul
2023-04-03 17:58:21 -04:00
джерело 4c9ed09270
коміт b2022d0869
3 змінених файлів з 119 додано та 11 видалено

Переглянути файл

@@ -1,4 +1,5 @@
ipython
num2words
omegaconf
pydub
PyYAML

Переглянути файл

@@ -1,4 +1,3 @@
import re
import time
from pathlib import Path
@@ -7,6 +6,8 @@ import modules.chat as chat
import modules.shared as shared
import torch
from extensions.silero_tts import tts_preprocessor
torch._C._jit_set_profiling_mode(False)
params = {
@@ -46,11 +47,6 @@ def load_model():
return model
model = load_model()
def remove_surrounded_chars(string):
# this expression matches to 'as few symbols as possible (0 upwards) between any asterisks' OR
# 'as few symbols as possible (0 upwards) between an asterisk and the end of the string'
return re.sub('\*[^\*]*?(\*|$)','',string)
def remove_tts_from_history(name1, name2):
for i, entry in enumerate(shared.history['internal']):
shared.history['visible'][i] = [shared.history['visible'][i][0], entry[1]]
@@ -98,11 +94,7 @@ def output_modifier(string):
return string
original_string = string
string = remove_surrounded_chars(string)
string = string.replace('"', '')
string = string.replace('', '')
string = string.replace('\n', ' ')
string = string.strip()
string = tts_preprocessor.preprocess(string)
if string == '':
string = '*Empty reply, try regenerating*'

115
extensions/silero_tts/tts_preprocessor.py Звичайний файл
Переглянути файл

@@ -0,0 +1,115 @@
import re
from num2words import num2words
alphabet_map = {
"A": " Ei ",
"B": " Bee ",
"C": " See ",
"D": " Dee ",
"E": " II ",
"F": " Eff ",
"G": " Jee ",
"H": " Eich ",
"I": " Eye ",
"J": " Jay ",
"K": " Kay ",
"L": " El ",
"M": " Emm ",
"N": " Enn ",
"O": " Ohh ",
"P": " Pii ",
"Q": " Queue ",
"R": " Are ",
"S": " Ess ",
"T": " Tee ",
"U": " You ",
"V": " Vii ",
"W": " Double You ",
"X": " Ex ",
"Y": " Why ",
"Z": "Zed" # Zed is weird, as I (da3dsoul) am American, but most of the voice models sound British, so it matches
}
def preprocess(string):
string = remove_surrounded_chars(string)
string = string.replace('"', '')
string = string.replace('', '')
string = string.replace('\n', ' ')
string = remove_commas(string)
string = hyphen_range_to(string)
string = num_to_words(string)
string = string.strip()
# TODO Try to use a ML predictor to expand abbreviations. It's hard, dependent on context, and whether to actually
# try to say the abbreviation or spell it out as I've done below is not agreed upon
# For now, expand abbreviations to pronunciations
string = replace_abbreviations(string)
return string
def replace_abbreviations(string):
pattern = re.compile(r'[\s("\'\[<][A-Z]{2,4}[\s,.?!)"\'\]>]')
result = string
while True:
match = pattern.search(result)
if match is None:
break
start = match.start()
end = match.end()
result = result[0:start] + replace_abbreviation(result[start:end]) + result[end:len(result)]
return result
def replace_abbreviation(string):
result = ""
for char in string:
result = match_mapping(char, result)
return result
def match_mapping(char, result):
for mapping in alphabet_map.keys():
if char == mapping:
return result + alphabet_map[char]
return result + char
def remove_surrounded_chars(string):
# this expression matches to 'as few symbols as possible (0 upwards) between any asterisks' OR
# 'as few symbols as possible (0 upwards) between an asterisk and the end of the string'
return re.sub(r'\*[^*]*?(\*|$)', '', string)
def hyphen_range_to(text):
pattern = re.compile(r'(\d+)[-](\d+)')
result = pattern.sub(lambda x: x.group(1) + ' to ' + x.group(2), text)
return result
def num_to_words(text):
pattern = re.compile(r'\d+')
result = pattern.sub(lambda x: num2words(int(x.group())), text)
return result
def remove_commas(text):
import re
pattern = re.compile(r'(\d),(\d)')
result = pattern.sub(r'\1\2', text)
return result
def __main__(args):
print(preprocess(args[1]))
if __name__ == "__main__":
import sys
__main__(sys.argv)