Spaces:
Runtime error
Runtime error
File size: 2,053 Bytes
a9482ab b87f08b 7f6563e 510f17f aeceb48 7f6563e 8772ca9 7f6563e d59ee2f 9630f4e dd29aa4 e28cac3 a9482ab a7d0893 9630f4e 55ef1e7 a9482ab 63ced49 a9482ab 7663e41 a9482ab 132c7ea 2b65d86 d2e0f91 b7d4e28 d2e0f91 b7d4e28 d2e0f91 dd29aa4 d9f9ad4 d2e0f91 91eda71 d9f9ad4 1256bad d9f9ad4 91eda71 7ad0327 91eda71 132c7ea 95ec227 91eda71 44e30b0 7670619 132c7ea 42ed2a2 d9f9ad4 a9482ab 2fce1e5 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 |
import IPython
import sys
import subprocess
subprocess.check_call([sys.executable, "-m", "pip", "install", "--upgrade", "--force-reinstall", "git+https://github.com/osanseviero/tortoise-tts.git"])
# entmax could not be installed at same time as torch
subprocess.check_call([sys.executable, "-m", "pip", "install", "entmax"])
from tortoise_tts.api import TextToSpeech
from tortoise_tts.utils.audio import load_audio, get_voices
import torch
import torchaudio
import gradio as gr
device = "cuda" if torch.cuda.is_available() else "cpu"
# This will download all the models used by Tortoise from the HF hub
tts = TextToSpeech(device="cuda")
voices = [
"angie",
"daniel",
"deniro",
"emma",
"freeman",
"geralt",
"halle",
"jlaw",
"lj",
"snakes",
"William",
]
voice_paths = get_voices()
print(voice_paths)
preset = "fast"
def inference(text, voice):
text = text[:256]
cond_paths = voice_paths[voice]
conds = []
print(voice_paths, voice, cond_paths)
for cond_path in cond_paths:
c = load_audio(cond_path, 22050)
conds.append(c)
print(text, conds, preset)
gen = tts.tts_with_preset(text, conds, preset)
print("gen")
torchaudio.save('generated.wav', gen.squeeze(0).cpu(), 24000)
return "generated.wav"
text = "Joining two modalities results in a surprising increase in generalization! What would happen if we combined them all?"
examples = [
[text, "angie"],
[text, "emma"],
["how are you doing this day", "freeman"]
]
iface = gr.Interface(
inference,
inputs=[
gr.inputs.Textbox(type="str", default=text, label="Text", lines=3),
gr.inputs.Dropdown(voices),
],
outputs="audio",
title="TorToiSe",
description="A multi-voice TTS system trained with an emphasis on quality",
article="This demo shows the ultra fast option in the TorToiSe system. For more info check the <a href='https://github.com/neonbjb/tortoise-tts' target='_blank'>Repository</a>.",
enable_queue=True,
examples=examples,
)
iface.launch(cache_examples=True)
|