re-enable redaction

This commit is contained in:
James Betker 2022-05-06 09:36:42 -06:00
parent 9151650559
commit e4e9523900
2 changed files with 2 additions and 2 deletions

View File

@ -165,7 +165,7 @@ class TextToSpeech:
Main entry point into Tortoise. Main entry point into Tortoise.
""" """
def __init__(self, autoregressive_batch_size=16, models_dir='.models', enable_redaction=False): def __init__(self, autoregressive_batch_size=16, models_dir='.models', enable_redaction=True):
""" """
Constructor Constructor
:param autoregressive_batch_size: Specifies how many samples to generate per batch. Lower this if you are seeing :param autoregressive_batch_size: Specifies how many samples to generate per batch. Lower this if you are seeing

View File

@ -50,7 +50,7 @@ class Wav2VecAlignment:
def __init__(self): def __init__(self):
self.model = Wav2Vec2ForCTC.from_pretrained("jbetker/wav2vec2-large-robust-ft-libritts-voxpopuli").cpu() self.model = Wav2Vec2ForCTC.from_pretrained("jbetker/wav2vec2-large-robust-ft-libritts-voxpopuli").cpu()
self.feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(f"facebook/wav2vec2-large-960h") self.feature_extractor = Wav2Vec2FeatureExtractor.from_pretrained(f"facebook/wav2vec2-large-960h")
self.tokenizer = Wav2Vec2CTCTokenizer.from_pretrained('jbetker/tacotron_symbols') self.tokenizer = Wav2Vec2CTCTokenizer.from_pretrained('jbetker/tacotron-symbols')
def align(self, audio, expected_text, audio_sample_rate=24000): def align(self, audio, expected_text, audio_sample_rate=24000):
orig_len = audio.shape[-1] orig_len = audio.shape[-1]