replace unidecode with anyascii

TTS/tts/utils/text/cleaners.py

@@ -1,18 +1,6 @@
-"""
-Cleaners are transformations that run over the input text at both training and eval time.
-
-Cleaners can be selected by passing a comma-delimited list of cleaner names as the "cleaners"
-hyperparameter. Some cleaners are English-specific. You'll typically want to use:
-  1. "english_cleaners" for English text
-  2. "transliteration_cleaners" for non-English text that can be transliterated to ASCII using
-     the Unidecode library (https://pypi.python.org/pypi/Unidecode)
-  3. "basic_cleaners" if you do not want to transliterate (in this case, you should also update
-     the symbols in symbols.py to match your data).
-"""
-
 import re
 
-from unidecode import unidecode
+from anyascii import anyascii
 
 from TTS.tts.utils.text.chinese_mandarin.numbers import replace_numbers_to_characters_in_text
 
@@ -47,7 +35,7 @@ def collapse_whitespace(text):
 
 
 def convert_to_ascii(text):
-    return unidecode(text)
+    return anyascii(text)
 
 
 def remove_aux_symbols(text):

requirements.txt

@@ -17,5 +17,5 @@ torch>=1.7
 tqdm
 numba==0.52
 umap-learn==0.4.6
-unidecode==0.4.20
+anyascii
 coqpit

tests/tts_tests/test_glow_tts_train.py

@@ -17,7 +17,7 @@ config = GlowTTSConfig(
     text_cleaner="english_cleaners",
     use_phonemes=True,
     phoneme_language="zh-CN",
-    phoneme_cache_path='tests/data/ljspeech/phoneme_cache/',
+    phoneme_cache_path="tests/data/ljspeech/phoneme_cache/",
     run_eval=True,
     test_delay_epochs=-1,
     epochs=1,

tests/tts_tests/test_speedy_speech_train.py

@@ -17,7 +17,7 @@ config = SpeedySpeechConfig(
     text_cleaner="english_cleaners",
     use_phonemes=True,
     phoneme_language="zh-CN",
-    phoneme_cache_path='tests/data/ljspeech/phoneme_cache/',
+    phoneme_cache_path="tests/data/ljspeech/phoneme_cache/",
     run_eval=True,
     test_delay_epochs=-1,
     epochs=1,

tests/vocoder_tests/test_melgan_train.py

@@ -19,6 +19,7 @@ config = MelganConfig(
     seq_len=2048,
     eval_split_size=1,
     print_step=1,
+    discriminator_model_params={"base_channels": 16, "max_channels": 256, "downsample_factors": [4, 4, 4]},
     print_eval=True,
     data_path="tests/data/ljspeech",
     output_path=output_path,