talipturkmen/tortoise-tts
1
0
Fork 0
forked from mrq/tortoise-tts
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix paths

Browse Source
This commit is contained in:
James Betker 2022-05-02 20:56:28 -06:00
parent e4e8ebfc55
commit 00e84bbd86
5 changed files with 6 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
.gitignore vendored
View File

@ -129,7 +129,6 @@ dmypy.json
.pyre/ .pyre/
.idea/* .idea/*
tortoise/.models/* .models/*
tortoise/random_voices/*
.custom/* .custom/*
results/* results/*

2
tortoise/models/arch_util.py
View File

@ -290,7 +290,7 @@ class AudioMiniEncoder(nn.Module):
class TorchMelSpectrogram(nn.Module): class TorchMelSpectrogram(nn.Module):
def __init__(self, filter_length=1024, hop_length=256, win_length=1024, n_mel_channels=80, mel_fmin=0, mel_fmax=8000, def __init__(self, filter_length=1024, hop_length=256, win_length=1024, n_mel_channels=80, mel_fmin=0, mel_fmax=8000,
sampling_rate=22050, normalize=False, mel_norm_file='data/mel_norms.pth'): sampling_rate=22050, normalize=False, mel_norm_file='tortoise/data/mel_norms.pth'):
super().__init__() super().__init__()
# These are the default tacotron values for the MEL spectrogram. # These are the default tacotron values for the MEL spectrogram.
self.filter_length = filter_length self.filter_length = filter_length

2
tortoise/read.py
View File

@ -28,7 +28,7 @@ def split_and_recombine_text(texts, desired_length=200, max_len=300):
if __name__ == '__main__': if __name__ == '__main__':
parser = argparse.ArgumentParser() parser = argparse.ArgumentParser()
parser.add_argument('--textfile', type=str, help='A file containing the text to read.', default="data/riding_hood.txt") parser.add_argument('--textfile', type=str, help='A file containing the text to read.', default="tortoise/data/riding_hood.txt")
parser.add_argument('--voice', type=str, help='Selects the voice to use for generation. See options in voices/ directory (and add your own!) ' parser.add_argument('--voice', type=str, help='Selects the voice to use for generation. See options in voices/ directory (and add your own!) '
'Use the & character to join two voices together. Use a comma to perform inference on multiple voices.', default='pat') 'Use the & character to join two voices together. Use a comma to perform inference on multiple voices.', default='pat')
parser.add_argument('--output_path', type=str, help='Where to store outputs.', default='../results/longform/') parser.add_argument('--output_path', type=str, help='Where to store outputs.', default='../results/longform/')

4
tortoise/utils/audio.py
View File

@ -82,10 +82,10 @@ def dynamic_range_decompression(x, C=1):
def get_voices(): def get_voices():
subs = os.listdir('voices') subs = os.listdir('tortoise/voices')
voices = {} voices = {}
for sub in subs: for sub in subs:
subj = os.path.join('voices', sub) subj = os.path.join('tortoise/voices', sub)
if os.path.isdir(subj): if os.path.isdir(subj):
voices[sub] = list(glob(f'{subj}/*.wav')) + list(glob(f'{subj}/*.mp3')) + list(glob(f'{subj}/*.pth')) voices[sub] = list(glob(f'{subj}/*.wav')) + list(glob(f'{subj}/*.mp3')) + list(glob(f'{subj}/*.pth'))
return voices return voices

2
tortoise/utils/tokenizer.py
View File

@ -164,7 +164,7 @@ def lev_distance(s1, s2):
return distances[-1] return distances[-1]
class VoiceBpeTokenizer: class VoiceBpeTokenizer:
def __init__(self, vocab_file='data/tokenizer.json'): def __init__(self, vocab_file='tortoise/data/tokenizer.json'):
if vocab_file is not None: if vocab_file is not None:
self.tokenizer = Tokenizer.from_file(vocab_file) self.tokenizer = Tokenizer.from_file(vocab_file)