Spaces:
Runtime error
Runtime error
File size: 4,075 Bytes
990357a fe48c10 4d74158 fe48c10 239000b fe48c10 990357a 6d066e2 990357a 6d066e2 f9e95cb 4d74158 8f9de4a 4d74158 f9e95cb 6d066e2 acda046 7fcb0ce f9e95cb 4a4f8b1 f9e95cb 6d066e2 7f29d20 acda046 f9e95cb 9243345 acda046 6d066e2 7fcb0ce |
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 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 |
import os
import gradio as gr
import random
os.system("pip install kantts -f https://modelscope.oss-cn-beijing.aliyuncs.com/releases/repo.html")
os.system("pip install numpy==1.22.0")
from modelscope.models.audio.tts import SambertHifigan
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks
# model_0
model_dir = os.path.abspath("./pretrain_work_dir")
custom_infer_abs = {
'voice_name':
'F7',
'am_ckpt':
os.path.join(model_dir, 'tmp_am', 'ckpt'),
'am_config':
os.path.join(model_dir, 'tmp_am', 'config.yaml'),
'voc_ckpt':
os.path.join(model_dir, 'orig_model', 'basemodel_16k', 'hifigan', 'ckpt'),
'voc_config':
os.path.join(model_dir, 'orig_model', 'basemodel_16k', 'hifigan',
'config.yaml'),
'audio_config':
os.path.join(model_dir, 'data', 'audio_config.yaml'),
'se_file':
os.path.join(model_dir, 'data', 'se', 'se.npy')
}
kwargs = {'custom_ckpt': custom_infer_abs}
model_id = SambertHifigan(os.path.join(model_dir, "orig_model"), **kwargs)
inference = pipeline(task=Tasks.text_to_speech, model=model_id)
# model_1
model_dir1 = os.path.abspath("./jay/pretrain_work_dir")
custom_infer_abs1 = {
'voice_name':
'F7',
'am_ckpt':
os.path.join(model_dir1, 'tmp_am', 'ckpt'),
'am_config':
os.path.join(model_dir1, 'tmp_am', 'config.yaml'),
'voc_ckpt':
os.path.join(model_dir1, 'orig_model', 'basemodel_16k', 'hifigan', 'ckpt'),
'voc_config':
os.path.join(model_dir1, 'orig_model', 'basemodel_16k', 'hifigan',
'config.yaml'),
'audio_config':
os.path.join(model_dir1, 'data', 'audio_config.yaml'),
'se_file':
os.path.join(model_dir1, 'data', 'se', 'se.npy')
}
kwargs1 = {'custom_ckpt': custom_infer_abs1}
model_id1 = SambertHifigan(os.path.join(model_dir1, "orig_model"), **kwargs1)
inference1 = pipeline(task=Tasks.text_to_speech, model=model_id1)
# functions
def infer(text):
output = inference(input=text)
filename = str(random.randint(1, 1000000000000))
with open(filename + "myfile.wav", mode='bx') as f:
f.write(output["output_wav"])
return filename + "myfile.wav"
def infer1(text):
output = inference1(input=text)
filename = str(random.randint(1, 1000000000000))
with open(filename + "file.wav", mode='bx') as f:
f.write(output["output_wav"])
return filename + "file.wav"
# upsample
import numpy as np
import torch
from hifi_gan_bwe import BandwidthExtender
MAX_LENGTH = 600.0
model = BandwidthExtender.from_pretrained("hifi-gan-bwe-10-42890e3-vctk-48kHz")
def extend(audio):
fs, x = audio
x = x[:int(MAX_LENGTH * fs)]
x = x.astype(np.float32) / 32767.0
if len(x.shape) == 1:
x = x[:, np.newaxis]
with torch.no_grad():
y = np.stack([model(torch.from_numpy(x), fs) for x in x.T]).T
y = (y * 32767.0).astype(np.int16)
fs = int(model.sample_rate)
return fs, y
app = gr.Blocks()
with app:
gr.HTML("<center>"
"<h1>🥳🎶🎡 - KanTTS中文声音克隆</h1>"
"</center>")
gr.Markdown("## <center>🌊 - 更多精彩应用，敬请关注[滔滔AI](http://www.talktalkai.com)；滔滔AI，为爱滔滔！💕</center>")
with gr.Row():
with gr.Column():
inp = gr.Textbox(lines=5, label="请填写您想要转换的中文文本")
with gr.Row():
btn = gr.Button("使用AI娜娜的声音", variant="primary")
btn1 = gr.Button("使用AI小杰的声音", variant="primary")
with gr.Column():
with gr.Row():
out = gr.Audio(label="为您生成的专属音频", type="filepath")
out1 = gr.Audio(label="更高采样率的专属音频", type="filepath")
btn2 = gr.Button("一键提高采样率")
btn.click(fn=infer, inputs=[inp], outputs=[out])
btn1.click(fn=infer1, inputs=[inp], outputs=[out])
btn2.click(fn=extend, inputs=[out], outputs=[out1])
app.launch(show_error=True) |