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

rename text to characters in config.json

This commit is contained in:
Edresson Casanova 2020-03-03 09:17:56 -03:00
parent 4e53896438
commit 36235c5e3f
13 changed files with 31 additions and 31 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
config.json
View File

@ -28,7 +28,7 @@
}, },
// VOCABULARY PARAMETERS // VOCABULARY PARAMETERS
"text":{ "characters":{
"pad": "_", "pad": "_",
"eos": "~", "eos": "~",
"bos": "^", "bos": "^",

4
notebooks/Benchmark-PWGAN.ipynb
View File

@ -143,8 +143,8 @@
" speaker_id = None\n", " speaker_id = None\n",
"\n", "\n",
"# if the vocabulary was passed, replace the default\n", "# if the vocabulary was passed, replace the default\n",
"if 'text' in CONFIG.keys():\n", "if 'characters' in CONFIG.keys():\n",
" symbols, phonemes = make_symbols(**CONFIG.text)\n", " symbols, phonemes = make_symbols(**CONFIG.characters)\n",
"\n", "\n",
"# load the model\n", "# load the model\n",
"num_chars = len(phonemes) if CONFIG.use_phonemes else len(symbols)\n", "num_chars = len(phonemes) if CONFIG.use_phonemes else len(symbols)\n",

4
notebooks/Benchmark.ipynb
View File

@ -150,8 +150,8 @@
" speaker_id = None\n", " speaker_id = None\n",
"\n", "\n",
"# if the vocabulary was passed, replace the default\n", "# if the vocabulary was passed, replace the default\n",
"if 'text' in CONFIG.keys():\n", "if 'characters' in CONFIG.keys():\n",
" symbols, phonemes = make_symbols(**CONFIG.text)\n", " symbols, phonemes = make_symbols(**CONFIG.characters)\n",
"\n", "\n",
"# load the model\n", "# load the model\n",
"num_chars = len(phonemes) if CONFIG.use_phonemes else len(symbols)\n", "num_chars = len(phonemes) if CONFIG.use_phonemes else len(symbols)\n",

6
notebooks/ExtractTTSpectrogram.ipynb
View File

@ -95,8 +95,8 @@
"outputs": [], "outputs": [],
"source": [ "source": [
"# if the vocabulary was passed, replace the default\n", "# if the vocabulary was passed, replace the default\n",
"if 'text' in C.keys():\n", "if 'characters' in C.keys():\n",
" symbols, phonemes = make_symbols(**C.text)\n", " symbols, phonemes = make_symbols(**C.characters)\n",
"\n", "\n",
"# load the model\n", "# load the model\n",
"num_chars = len(phonemes) if C.use_phonemes else len(symbols)\n", "num_chars = len(phonemes) if C.use_phonemes else len(symbols)\n",
@ -120,7 +120,7 @@
"preprocessor = importlib.import_module('TTS.datasets.preprocess')\n", "preprocessor = importlib.import_module('TTS.datasets.preprocess')\n",
"preprocessor = getattr(preprocessor, DATASET.lower())\n", "preprocessor = getattr(preprocessor, DATASET.lower())\n",
"meta_data = preprocessor(DATA_PATH,METADATA_FILE)\n", "meta_data = preprocessor(DATA_PATH,METADATA_FILE)\n",
"dataset = MyDataset(checkpoint['r'], C.text_cleaner, ap, meta_data,tp=C.text if 'text' in C.keys() else None, use_phonemes=C.use_phonemes, phoneme_cache_path=C.phoneme_cache_path, enable_eos_bos=C.enable_eos_bos_chars)\n", "dataset = MyDataset(checkpoint['r'], C.text_cleaner, ap, meta_data,tp=C.characters if 'characters' in C.keys() else None, use_phonemes=C.use_phonemes, phoneme_cache_path=C.phoneme_cache_path, enable_eos_bos=C.enable_eos_bos_chars)\n",
"loader = DataLoader(dataset, batch_size=BATCH_SIZE, num_workers=4, collate_fn=dataset.collate_fn, shuffle=False, drop_last=False)" "loader = DataLoader(dataset, batch_size=BATCH_SIZE, num_workers=4, collate_fn=dataset.collate_fn, shuffle=False, drop_last=False)"
] ]
}, },

4
notebooks/TestAttention.ipynb
View File

@ -111,8 +111,8 @@
" speaker_id = None\n", " speaker_id = None\n",
"\n", "\n",
"# if the vocabulary was passed, replace the default\n", "# if the vocabulary was passed, replace the default\n",
"if 'text' in CONFIG.keys():\n", "if 'characters' in CONFIG.keys():\n",
" symbols, phonemes = make_symbols(**CONFIG.text)\n", " symbols, phonemes = make_symbols(**CONFIG.characters)\n",
"\n", "\n",
"# load the model\n", "# load the model\n",
"num_chars = len(phonemes) if CONFIG.use_phonemes else len(symbols)\n", "num_chars = len(phonemes) if CONFIG.use_phonemes else len(symbols)\n",

4
server/synthesizer.py
View File

@ -52,8 +52,8 @@ class Synthesizer(object):
self.use_phonemes = self.tts_config.use_phonemes self.use_phonemes = self.tts_config.use_phonemes
self.ap = AudioProcessor(**self.tts_config.audio) self.ap = AudioProcessor(**self.tts_config.audio)
if 'text' in self.tts_config.keys(): if 'characters' in self.tts_config.keys():
symbols, phonemes = make_symbols(**self.tts_config.text) symbols, phonemes = make_symbols(**self.tts_config.characters)
if self.use_phonemes: if self.use_phonemes:
self.input_size = len(phonemes) self.input_size = len(phonemes)

4
synthesize.py
View File

@ -108,8 +108,8 @@ if __name__ == "__main__":
ap = AudioProcessor(**C.audio) ap = AudioProcessor(**C.audio)
# if the vocabulary was passed, replace the default # if the vocabulary was passed, replace the default
if 'text' in C.keys(): if 'characters' in C.keys():
symbols, phonemes = make_symbols(**C.text) symbols, phonemes = make_symbols(**C.characters)
# load speakers # load speakers
if args.speakers_json != '': if args.speakers_json != '':

4
tests/test_demo_server.py
View File

@ -15,8 +15,8 @@ class DemoServerTest(unittest.TestCase):
# pylint: disable=global-statement # pylint: disable=global-statement
global symbols, phonemes global symbols, phonemes
config = load_config(os.path.join(get_tests_output_path(), 'dummy_model_config.json')) config = load_config(os.path.join(get_tests_output_path(), 'dummy_model_config.json'))
if 'text' in config.keys(): if 'characters' in config.keys():
symbols, phonemes = make_symbols(**config.text) symbols, phonemes = make_symbols(**config.characters)
num_chars = len(phonemes) if config.use_phonemes else len(symbols) num_chars = len(phonemes) if config.use_phonemes else len(symbols)
model = setup_model(num_chars, 0, config) model = setup_model(num_chars, 0, config)

2
tests/test_loader.py
View File

@ -38,7 +38,7 @@ class TestTTSDataset(unittest.TestCase):
c.text_cleaner, c.text_cleaner,
ap=self.ap, ap=self.ap,
meta_data=items, meta_data=items,
tp=c.text if 'text' in c.keys() else None, tp=c.characters if 'characters' in c.keys() else None,
batch_group_size=bgs, batch_group_size=bgs,
min_seq_len=c.min_seq_len, min_seq_len=c.min_seq_len,
max_seq_len=float("inf"), max_seq_len=float("inf"),

6
train.py
View File

@ -49,7 +49,7 @@ def setup_loader(ap, r, is_val=False, verbose=False):
c.text_cleaner, c.text_cleaner,
meta_data=meta_data_eval if is_val else meta_data_train, meta_data=meta_data_eval if is_val else meta_data_train,
ap=ap, ap=ap,
tp=c.text if 'text' in c.keys() else None, tp=c.characters if 'characters' in c.keys() else None,
batch_group_size=0 if is_val else c.batch_group_size * batch_group_size=0 if is_val else c.batch_group_size *
c.batch_size, c.batch_size,
min_seq_len=c.min_seq_len, min_seq_len=c.min_seq_len,
@ -520,8 +520,8 @@ def main(args): # pylint: disable=redefined-outer-name
global meta_data_train, meta_data_eval, symbols, phonemes global meta_data_train, meta_data_eval, symbols, phonemes
# Audio processor # Audio processor
ap = AudioProcessor(**c.audio) ap = AudioProcessor(**c.audio)
if 'text' in c.keys(): if 'characters' in c.keys():
symbols, phonemes = make_symbols(**c.text) symbols, phonemes = make_symbols(**c.characters)
# DISTRUBUTED # DISTRUBUTED
if num_gpus > 1: if num_gpus > 1:

14
utils/generic_utils.py
View File

@ -426,13 +426,13 @@ def check_config(c):
_check_argument('griffin_lim_iters', c['audio'], restricted=True, val_type=int, min_val=10, max_val=1000) _check_argument('griffin_lim_iters', c['audio'], restricted=True, val_type=int, min_val=10, max_val=1000)
# vocabulary parameters # vocabulary parameters
_check_argument('text', c, restricted=False, val_type=dict) _check_argument('characters', c, restricted=False, val_type=dict)
_check_argument('pad', c['text'] if 'text' in c.keys() else {}, restricted='text' in c.keys(), val_type=str) _check_argument('pad', c['characters'] if 'characters' in c.keys() else {}, restricted='characters' in c.keys(), val_type=str)
_check_argument('eos', c['text'] if 'text' in c.keys() else {}, restricted='text' in c.keys(), val_type=str) _check_argument('eos', c['characters'] if 'characters' in c.keys() else {}, restricted='characters' in c.keys(), val_type=str)
_check_argument('bos', c['text'] if 'text' in c.keys() else {}, restricted='text' in c.keys(), val_type=str) _check_argument('bos', c['characters'] if 'characters' in c.keys() else {}, restricted='characters' in c.keys(), val_type=str)
_check_argument('characters', c['text'] if 'text' in c.keys() else {}, restricted='text' in c.keys(), val_type=str) _check_argument('characters', c['characters'] if 'characters' in c.keys() else {}, restricted='characters' in c.keys(), val_type=str)
_check_argument('phonemes', c['text'] if 'text' in c.keys() else {}, restricted='text' in c.keys(), val_type=str) _check_argument('phonemes', c['characters'] if 'characters' in c.keys() else {}, restricted='characters' in c.keys(), val_type=str)
_check_argument('punctuations', c['text'] if 'text' in c.keys() else {}, restricted='text' in c.keys(), val_type=str) _check_argument('punctuations', c['characters'] if 'characters' in c.keys() else {}, restricted='characters' in c.keys(), val_type=str)
# normalization parameters # normalization parameters
_check_argument('signal_norm', c['audio'], restricted=True, val_type=bool) _check_argument('signal_norm', c['audio'], restricted=True, val_type=bool)

4
utils/synthesis.py
View File

@ -10,10 +10,10 @@ def text_to_seqvec(text, CONFIG, use_cuda):
seq = np.asarray( seq = np.asarray(
phoneme_to_sequence(text, text_cleaner, CONFIG.phoneme_language, phoneme_to_sequence(text, text_cleaner, CONFIG.phoneme_language,
CONFIG.enable_eos_bos_chars, CONFIG.enable_eos_bos_chars,
tp=CONFIG.text if 'text' in CONFIG.keys() else None), tp=CONFIG.characters if 'characters' in CONFIG.keys() else None),
dtype=np.int32) dtype=np.int32)
else: else:
seq = np.asarray(text_to_sequence(text, text_cleaner, tp=CONFIG.text if 'text' in CONFIG.keys() else None), dtype=np.int32) seq = np.asarray(text_to_sequence(text, text_cleaner, tp=CONFIG.characters if 'characters' in CONFIG.keys() else None), dtype=np.int32)
# torch tensor # torch tensor
chars_var = torch.from_numpy(seq).unsqueeze(0) chars_var = torch.from_numpy(seq).unsqueeze(0)
if use_cuda: if use_cuda:

4
utils/visual.py
View File

@ -54,8 +54,8 @@ def visualize(alignment, spectrogram_postnet, stop_tokens, text, hop_length, CON
plt.xlabel("Decoder timestamp", fontsize=label_fontsize) plt.xlabel("Decoder timestamp", fontsize=label_fontsize)
plt.ylabel("Encoder timestamp", fontsize=label_fontsize) plt.ylabel("Encoder timestamp", fontsize=label_fontsize)
if CONFIG.use_phonemes: if CONFIG.use_phonemes:
seq = phoneme_to_sequence(text, [CONFIG.text_cleaner], CONFIG.phoneme_language, CONFIG.enable_eos_bos_chars, tp=CONFIG.text if 'text' in CONFIG.keys() else None) seq = phoneme_to_sequence(text, [CONFIG.text_cleaner], CONFIG.phoneme_language, CONFIG.enable_eos_bos_chars, tp=CONFIG.characters if 'characters' in CONFIG.keys() else None)
text = sequence_to_phoneme(seq, tp=CONFIG.text if 'text' in CONFIG.keys() else None) text = sequence_to_phoneme(seq, tp=CONFIG.characters if 'characters' in CONFIG.keys() else None)
print(text) print(text)
plt.yticks(range(len(text)), list(text)) plt.yticks(range(len(text)), list(text))