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

Fix tests

This commit is contained in:
WeberJulian 2021-11-23 16:00:38 +01:00 committed by Eren Gölge
parent 78a23e19df
commit 09eda31a3f
3 changed files with 11 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
TTS/tts/models/vits.py
View File

@ -596,7 +596,7 @@ class Vits(BaseTTS):
# language embedding
lang_emb = None
if hasattr(self, "emb_l"):
if self.args.use_language_embedding and lid is not None:
lang_emb = self.emb_l(lid).unsqueeze(-1)
x, m_p, logs_p, x_mask = self.text_encoder(x, x_lengths, lang_emb=lang_emb)

10
tests/tts_tests/test_vits_multilingual_train-d_vectors.py
View File

@ -10,7 +10,7 @@ config_path = os.path.join(get_tests_output_path(), "test_model_config.json")
output_path = os.path.join(get_tests_output_path(), "train_outputs")
dataset_config1 = BaseDatasetConfig(
dataset_config_en = BaseDatasetConfig(
name="ljspeech",
meta_file_train="metadata.csv",
meta_file_val="metadata.csv",
@ -18,12 +18,12 @@ dataset_config1 = BaseDatasetConfig(
language="en",
)
dataset_config2 = BaseDatasetConfig(
dataset_config_pt = BaseDatasetConfig(
name="ljspeech",
meta_file_train="metadata.csv",
meta_file_val="metadata.csv",
path="tests/data/ljspeech",
language="en2",
language="pt-br",
)
config = VitsConfig(
@ -43,9 +43,9 @@ config = VitsConfig(
print_eval=True,
test_sentences=[
["Be a voice, not an echo.", "ljspeech-0", None, "en"],
["Be a voice, not an echo.", "ljspeech-1", None, "en2"],
["Be a voice, not an echo.", "ljspeech-1", None, "pt-br"],
],
datasets=[dataset_config1, dataset_config2],
datasets=[dataset_config_en, dataset_config_pt],
)
# set audio config
config.audio.do_trim_silence = True

10
tests/tts_tests/test_vits_multilingual_train.py
View File

@ -10,7 +10,7 @@ config_path = os.path.join(get_tests_output_path(), "test_model_config.json")
output_path = os.path.join(get_tests_output_path(), "train_outputs")
dataset_config1 = BaseDatasetConfig(
dataset_config_en = BaseDatasetConfig(
name="ljspeech",
meta_file_train="metadata.csv",
meta_file_val="metadata.csv",
@ -18,12 +18,12 @@ dataset_config1 = BaseDatasetConfig(
language="en",
)
dataset_config2 = BaseDatasetConfig(
dataset_config_pt = BaseDatasetConfig(
name="ljspeech",
meta_file_train="metadata.csv",
meta_file_val="metadata.csv",
path="tests/data/ljspeech",
language="en2",
language="pt-br",
)
config = VitsConfig(
@ -43,9 +43,9 @@ config = VitsConfig(
print_eval=True,
test_sentences=[
["Be a voice, not an echo.", "ljspeech", None, "en"],
["Be a voice, not an echo.", "ljspeech", None, "en2"],
["Be a voice, not an echo.", "ljspeech", None, "pt-br"],
],
datasets=[dataset_config1, dataset_config2],
datasets=[dataset_config_en, dataset_config_pt],
)
# set audio config
config.audio.do_trim_silence = True