File size: 2,849 Bytes
8917cb5
 
 
 
 
 
 
 
 
 
 
 
faa52ce
 
8917cb5
 
 
275d258
8917cb5
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
275d258
faa52ce
 
 
8917cb5
 
 
faa52ce
 
 
 
 
 
ed044ca
8917cb5
 
 
 
 
 
 
 
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
import io

import gradio as gr
import librosa
import numpy as np
import soundfile
import torch
from inference.infer_tool import Svc
import logging

logging.getLogger('numba').setLevel(logging.WARNING)

model_name = "logs/32k/tiehu.pth"
config_name = "configs/tiehu.json"

svc_model = Svc(model_name, config_name)
sid_map = {
    "南云铁虎": "tiehu"
}


def vc_fn(sid, input_audio, vc_transform):
    if input_audio is None:
        return "You need to upload an audio", None
    sampling_rate, audio = input_audio
    # print(audio.shape,sampling_rate)
    duration = audio.shape[0] / sampling_rate
    if duration > 45:
        return "请上传小于45s的音频,需要转换长音频请本地进行转换", None
    audio = (audio / np.iinfo(audio.dtype).max).astype(np.float32)
    if len(audio.shape) > 1:
        audio = librosa.to_mono(audio.transpose(1, 0))
    if sampling_rate != 16000:
        audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=16000)
    print(audio.shape)
    out_wav_path = io.BytesIO()
    soundfile.write(out_wav_path, audio, 16000, format="wav")
    out_wav_path.seek(0)

    sid = sid_map[sid]
    out_audio, out_sr = svc_model.infer(sid, vc_transform, out_wav_path)
    _audio = out_audio.cpu().numpy()
    return "Success", (32000, _audio)


app = gr.Blocks()
with app:
    with gr.Tabs():
        with gr.TabItem("Basic"):
            gr.Markdown(value="""
               
                南云铁虎Sovits3.0模型
                如需本地使用,下载files里configs文件夹中.json格式文件,
                                     logs/32k文件夹中.pth格式文件
                                     3.0.zip文件并解压,查看说明.txt文件
                
                项目改写基于 https://huggingface.co/spaces/innnky/nyaru-svc-3.0
                
                                模型使用协议(重要):
                1.请勿用于商业目的
                2.请勿用于会影响本人的行为(比如冒充本人发表争议言论)
                3.请勿用于血腥、暴力、性相关、政治相关内容
                4.非个人使用场合请注明模型作者
                5.允许用于个人娱乐场景下的游戏语音、直播活动
            sid = gr.Dropdown(label="音色", choices=["南云铁虎"], value="tiehu")
            vc_input3 = gr.Audio(label="上传音频(长度小于45秒)")
            vc_transform = gr.Number(label="变调(整数,可以正负,半音数量,升高八度就是12)", value=0)
            vc_submit = gr.Button("转换", variant="primary")
            vc_output1 = gr.Textbox(label="Output Message")
            vc_output2 = gr.Audio(label="Output Audio")
        vc_submit.click(vc_fn, [sid, vc_input3, vc_transform], [vc_output1, vc_output2])

    app.launch()