dianskip/tortoise-tts
1
0
Fork 0
forked from mrq/tortoise-tts
Code Pull Requests Activity

fixed using old output dir because of my autism with prefixing everything with "./" broke it, fixed incrementing filenames

Browse Source
This commit is contained in:
mrq 2023-02-11 12:39:16 +00:00
parent 58e2b22b0e
commit 44eba62dc8
1 changed files with 15 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
webui.py
View File

@ -14,18 +14,12 @@ import gradio.utils
from datetime import datetime from datetime import datetime
from fastapi import FastAPI
import tortoise.api import tortoise.api
from tortoise.api import TextToSpeech from tortoise.api import TextToSpeech
from tortoise.utils.audio import load_audio, load_voice, load_voices, get_voice_dir from tortoise.utils.audio import load_audio, load_voice, load_voices, get_voice_dir
from tortoise.utils.text import split_and_recombine_text from tortoise.utils.text import split_and_recombine_text
args = None
webui = None
tts = None
def generate( def generate(
text, text,
delimiter, delimiter,
@ -49,6 +43,9 @@ def generate(
experimental_checkboxes, experimental_checkboxes,
progress=gr.Progress(track_tqdm=True) progress=gr.Progress(track_tqdm=True)
): ):
global args
global tts
try: try:
tts tts
except NameError: except NameError:
@ -77,7 +74,7 @@ def generate(
conditioning_latents = (conditioning_latents[0], conditioning_latents[1], conditioning_latents[2], None) conditioning_latents = (conditioning_latents[0], conditioning_latents[1], conditioning_latents[2], None)
if voice != "microphone": if voice != "microphone":
torch.save(conditioning_latents, f'./{get_voice_dir()}/{voice}/cond_latents.pth') torch.save(conditioning_latents, f'{get_voice_dir()}/{voice}/cond_latents.pth')
voice_samples = None voice_samples = None
else: else:
sample_voice = None sample_voice = None
@ -146,6 +143,8 @@ def generate(
for i, file in enumerate(os.listdir(outdir)): for i, file in enumerate(os.listdir(outdir)):
if file[-5:] == ".json": if file[-5:] == ".json":
idx = idx + 1 idx = idx + 1
if idx:
idx = idx + 1
def get_name(line=0, candidate=0, combined=False): def get_name(line=0, candidate=0, combined=False):
if combined: if combined:
@ -260,7 +259,7 @@ def generate(
f.write(json.dumps(info, indent='\t') ) f.write(json.dumps(info, indent='\t') )
if voice is not None and conditioning_latents is not None: if voice is not None and conditioning_latents is not None:
with open(f'./{get_voice_dir()}/{voice}/cond_latents.pth', 'rb') as f: with open(f'{get_voice_dir()}/{voice}/cond_latents.pth', 'rb') as f:
info['latents'] = base64.b64encode(f.read()).decode("ascii") info['latents'] = base64.b64encode(f.read()).decode("ascii")
if args.embed_output_metadata: if args.embed_output_metadata:
@ -277,7 +276,10 @@ def generate(
if sample_voice is not None: if sample_voice is not None:
sample_voice = (tts.input_sample_rate, sample_voice.numpy()) sample_voice = (tts.input_sample_rate, sample_voice.numpy())
print(f"Generation took {info['time']} seconds, saved to '{outdir}'\n") if output_voice is None and len(output_voices):
output_voice = output_voices[0]
print(f"Generation took {info['time']} seconds, saved to '{output_voice}'\n")
info['seed'] = settings['use_deterministic_seed'] info['seed'] = settings['use_deterministic_seed']
del info['latents'] del info['latents']
@ -290,7 +292,7 @@ def generate(
return ( return (
sample_voice, sample_voice,
output_voice if output_voice is not None else output_voices[0], output_voice,
results, results,
) )
@ -329,7 +331,7 @@ def read_generate_settings(file, save_latents=True, save_as_temp=True):
del j['latents'] del j['latents']
if latents and save_latents: if latents and save_latents:
outdir=f'./{get_voice_dir()}/{".temp" if save_as_temp else j["voice"]}/' outdir=f'{get_voice_dir()}/{".temp" if save_as_temp else j["voice"]}/'
os.makedirs(outdir, exist_ok=True) os.makedirs(outdir, exist_ok=True)
with open(f'{outdir}/cond_latents.pth', 'wb') as f: with open(f'{outdir}/cond_latents.pth', 'wb') as f:
f.write(latents) f.write(latents)
@ -517,12 +519,14 @@ def setup_args():
return args return args
def setup_tortoise(): def setup_tortoise():
global args
print("Initializating TorToiSe...") print("Initializating TorToiSe...")
tts = TextToSpeech(minor_optimizations=not args.low_vram) tts = TextToSpeech(minor_optimizations=not args.low_vram)
print("TorToiSe initialized, ready for generation.") print("TorToiSe initialized, ready for generation.")
return tts return tts
def setup_gradio(): def setup_gradio():
global args
if not args.share: if not args.share:
def noop(function, return_value=None): def noop(function, return_value=None):
def wrapped(*args, **kwargs): def wrapped(*args, **kwargs):