Hugging Face's logo

Spaces:
oceannnn
/
Modelscope_Faster_Whisper_Multi_Subtitle
Runtime error

App Files Community
Modelscope_Faster_Whisper_Multi_Subtitle
File size: 4,789 Bytes 
cb0791d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
 
import argparse
import os

import gradio as gr

from utils import movie2audio,make_srt,make_tran,merge_sub,make_tran_zh2en,make_tran_ja2zh,make_tran_ko2zh




initial_md = """


作者：刘悦的技术博客  https://space.bilibili.com/3031494

"""


def do_speech(video):

    res = movie2audio(video)

    return res


def do_trans_video(model_type,video_path):

    srt_text = make_srt(video_path,model_type)

    return srt_text

def do_trans_audio(model_type):

    srt_text = make_srt('./audio.wav',model_type)

    return srt_text

def do_trans_en2zh(srt_path):

    return make_tran(srt_path)

def do_trans_zh2en(srt_path):

    return make_tran_zh2en(srt_path)

def do_trans_ja2zh(srt_path):

    return make_tran_ja2zh(srt_path)

def do_trans_ko2zh(srt_path):

    return make_tran_ko2zh(srt_path)

def do_srt_sin(video_path):

    return merge_sub(video_path,"./video.srt")

def do_srt_two(video_path):

    return merge_sub(video_path,"./two.srt")



with gr.Blocks() as app:
    gr.Markdown(initial_md)

    with gr.Accordion("视频处理(Video)"):
        with gr.Row():

            ori_video = gr.Textbox(label="请输入视频的路径地址，如:d:/123.mp4")
        
            speech_button = gr.Button("提取人声(如果视频没有背景音也可以不做)Extract human voice (you don't have to do it if the video has no background sound)")

            speech_audio = gr.Audio(label="提取的人声(Extract voice)")

    
    speech_button.click(do_speech,inputs=[ori_video],outputs=[speech_audio])
    
    with gr.Accordion("转写字幕"):
        with gr.Row():
            with gr.Column():
                

                # model_type = gr.Dropdown(choices=["small","medium","large-v3","large-v2"], value="small", label="选择faster_Whisper模型/Select faster_Whisper model",interactive=True)

                model_type = gr.Textbox(label="填写faster_Whisper模型/Fill in the faster_Whisper model,也可以填写small,medium,large,large-v2,large-v3,模型越大，速度越慢，但字幕的准确度越高，酌情填写，用文本框是因为你可以填写其他huggingface上的开源模型地址",value="medium")

                
            transcribe_button_whisper = gr.Button("视频直接转写字幕(Video direct rewriting subtitles)")

            transcribe_button_audio = gr.Button("提取人声转写字幕(Extract voice transliteration subtitles)")

            result1 = gr.Textbox(label="字幕結果(会在项目目录生成video.srt/video.srt is generated in the current directory)")

        transcribe_button_whisper.click(do_trans_video,inputs=[model_type,ori_video],outputs=[result1])

        transcribe_button_audio.click(do_trans_audio,inputs=[model_type],outputs=[result1])


    with gr.Accordion("字幕翻译"):
        with gr.Row():


            srt_path = gr.Textbox(label="原始字幕地址，默认为项目目录中的video.srt,也可以输入其他路径",value="./video.srt")

            trans_button_en2zh = gr.Button("翻译英语字幕为中文/Translate English subtitles into Chinese")

            trans_button_zh2en = gr.Button("翻译中文字幕为英文/Translate Chinese subtitles into English")

            trans_button_ja2zh = gr.Button("翻译日文字幕为中文/Translate Japanese subtitles into Chinese")

            trans_button_ko2zh = gr.Button("翻译韩文字幕为中文/Translate Korea subtitles into Chinese")

            result2 = gr.Textbox(label="翻译结果(会在项目目录生成two.srt/two.srt is generated in the current directory)")

        trans_button_en2zh.click(do_trans_en2zh,[srt_path],outputs=[result2])

        trans_button_zh2en.click(do_trans_zh2en,[srt_path],outputs=[result2])

        trans_button_ja2zh.click(do_trans_ja2zh,[srt_path],outputs=[result2])

        trans_button_ko2zh.click(do_trans_ko2zh,[srt_path],outputs=[result2])

    with gr.Accordion("字幕合并"):
        with gr.Row():


            srt_button_sin = gr.Button("将单语字幕合并到视频/Merge monolingual subtitles into video")

            srt_button_two = gr.Button("将双语字幕合并到视频/Merge bilingual subtitles into video")

            result3 = gr.Textbox(label="合成字幕后的视频路径地址")

    srt_button_sin.click(do_srt_sin,inputs=[ori_video],outputs=[result3])
    srt_button_two.click(do_srt_two,inputs=[ori_video],outputs=[result3])


    

parser = argparse.ArgumentParser()
parser.add_argument(
    "--server-name",
    type=str,
    default=None,
    help="Server name for Gradio app",
)
parser.add_argument(
    "--no-autolaunch",
    action="store_true",
    default=False,
    help="Do not launch app automatically",
)
args = parser.parse_args()

app.launch(inbrowser=not args.no_autolaunch, server_name=args.server_name)