Skip to content

Commit f4925f3

Browse files
committed
更新VSE v2.0.0
1 parent b25b2d8 commit f4925f3

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

57 files changed

+293
-220
lines changed

README.md

+34-18
Original file line numberDiff line numberDiff line change
@@ -12,14 +12,15 @@ Video-subtitle-extractor (vse) 是一款将视频中的硬字幕提取为外挂
1212
- 提取视频中的关键帧
1313
- 检测视频帧中文本的所在位置
1414
- 识别视频帧中文本的内容
15-
- 过滤非字幕区域的文本,去除水印(台标)文本
15+
- 过滤非字幕区域的文本
16+
- [去除水印、台标文本、原视频硬字幕,可配合:video-subtitle-remover (vsr) ](https://github.com/YaoFANGUK/video-subtitle-remover/tree/main)
1617
- 去除重复字幕行,生成srt字幕文件
1718
- 支持视频字幕**批量提取**(打开文件的时候选择多个视频)
18-
- 多语言:支持**中文/英文****日语****韩语****阿拉伯语****繁体中文****法语****德语****俄语****西班牙语****葡萄牙语****意大利语**字幕的提取
19+
- 多语言:支持**简体中文(中英双语)****繁体中文****英文****日语****韩语****越南语****阿拉伯语****法语****德语****俄语****西班牙语****葡萄牙语****意大利语****87种**语言的字幕提取
1920
- 多模式:
20-
- **快速** - 快速提取字幕但可能丢字幕(推荐)
21-
- **精准** - 不丢字幕但速度较慢
22-
21+
- **快速**:使用轻量模型,快速提取字幕,但可能丢字幕(推荐)
22+
- **自动**:自动判断模型,CPU下使用轻量模型;GPU下使用精准模型,提取字幕速度较慢,几乎不丢字幕
23+
- **精准**:使用精准模型,GPU下逐帧检测,不丢字幕,但速度**非常慢**(非常不推荐)
2324

2425
**使用说明**
2526

@@ -37,15 +38,19 @@ Video-subtitle-extractor (vse) 是一款将视频中的硬字幕提取为外挂
3738

3839
**下载地址**
3940

40-
- Windows 单文件版本: <a href=https://www.aliyundrive.com/s/7mdRDJHVRiu target=_blank>vse.exe</a> 提取码:**17uj**
41+
- Windows 绿色版本v2.0.0(CPU): <a href="https://pan.baidu.com/s/1aUtZqGix1J0aqwGX4VRCWA?pwd=vse2" target=_blank>vse_windows_cpu_v2.0.0.zip</a> 提取码:**vse2**
42+
43+
> **推荐使用,启动速度较快**
4144
42-
> (双击直接运行,每次打开时会有一点慢,**推荐小白使用**)
45+
- Windows 单文件版本v2.0.0(CPU): <a href=https://www.aliyundrive.com/s/uD5ZfoAbDcf target=_blank>vse.exe</a> 提取码:**rl02**
4346

44-
- Windows GPU版本: <a href="https://pan.baidu.com/s/1L-cdVHnBGva2bGrsStYnnQ">vse_windows_GPU.7z</a> 提取码:**r911**
47+
> 双击直接运行,每次打开时会有一点慢,**若出现误报毒,使用绿色版**
4548
46-
- Windows CPU版本: <a href="https://pan.baidu.com/s/1oPyJowF0XgHtgg2g9EA-nw">vse_windows_CPU.zip</a> 提取码:**1vi6**
49+
- Windows GPU版本v2.0.0(GPU): <a href="https://pan.baidu.com/s/1w8vJnqF91RlJB39LMbZmDg?pwd=vse2">vse_windows_gpu_v2.0.0.7z</a> 提取码:**vse2**
4750

48-
- MacOS CPU版本: <a href="https://pan.baidu.com/s/1WgZpr_8I3Dv7A8ThwcIPng">vse_macOS_CPU.dmg</a> 提取码:**7gbo**
51+
> **仅供具有Nvidia显卡的用户使用(AMD的显卡不行),提取速度非常快**
52+
53+
- MacOS 版本v0.1.0(CPU): <a href="https://pan.baidu.com/s/1WgZpr_8I3Dv7A8ThwcIPng">vse_macOS_CPU.dmg</a> 提取码:**7gbo**
4954

5055
> PS: 若无法下载,请前往<a href="https://github.com/YaoFANGUK/video-subtitle-extractor/releases"> Release </a>下载
5156
@@ -57,7 +62,8 @@ Video-subtitle-extractor (vse) 是一款将视频中的硬字幕提取为外挂
5762
- (CLI版本) 无需用户手动设置字幕区域,项目通过文本检测模型自动检测字幕区域
5863
- (GUI版本) 图形化界面
5964

60-
<img src="https://z3.ax1x.com/2021/04/09/cNrA1A.png" alt="demo">
65+
<p style="text-align:center;"><img src="https://github.com/YaoFANGUK/video-subtitle-extractor/raw/main/design/demo.png" alt="demo.png"/></p>
66+
6167

6268
点击【打开】后选择视频文件,调整字幕区域,点击【运行】
6369

@@ -225,7 +231,7 @@ python ./backend/main.py
225231

226232
将项目中的.condarc放在用户目录下(C:\Users\\<你的用户名>),如果用户目录已经存在该文件则覆盖
227233

228-
解决方案:<a href="https://zhuanlan.zhihu.com/p/260034241">https://zhuanlan.zhihu.com/p/260034241</a>
234+
解决方案:<a href="https://zhuanlan.zhihu.com/p/260034241">https://zhuanlan.zhihu.com/p/260034241 </a>
229235

230236
#### 3. Windows下出现geos_c.dll错误
231237

@@ -238,20 +244,23 @@ OSError: [WinError 126] 找不到指定的模块。
238244

239245
解决方案:
240246

241-
1) 卸载Shapely
247+
(1) 卸载Shapely
242248

243249
```shell
244250
pip uninstall Shapely -y
245251
```
246252

247-
2) 使用conda重新安装Shapely
253+
(2) 使用conda重新安装Shapely
248254

249255
```shell
250256
conda install Shapely
251257
```
252258

259+
#### 4. 7z文件解压错误
253260

254-
#### 4. Nuitka打包代码闪退
261+
解决方案:升级7-zip解压程序到最新版本
262+
263+
#### 5. Nuitka打包代码闪退
255264

256265
使用Nuitka版本```0.6.19```,将conda虚拟环境Lib文件夹下site-packages的所有文件复制到dependencies文件夹中,把paddle库dataset下image.py的有关subprocess代码全部注释了,使用以下打包命令:
257266

@@ -261,11 +270,9 @@ conda install Shapely
261270

262271
编译成单个文件(pip安装zstandard可以减小体积)
263272
```shell
264-
python -m nuitka --standalone --windows-disable-console --mingw64 --lto no --include-data-dir=D:\vse\backend=backend --include-data-dir=D:\vse\dependencies=dependencies --nofollow-imports --windows-icon-from-ico=D:\vse\design\vse.ico --plugin-enable=tk-inter,multiprocessing --output-dir=out --onefile .\gui.py
273+
python -m nuitka --standalone --windows-disable-console --mingw64 --lto no --include-data-dir=C:\Users\Yao\Downloads\vse\backend=backend --include-data-dir=C:\Users\Yao\Downloads\vse\design=design --include-data-dir=C:\Users\Yao\Downloads\vse\dependencies=dependencies --nofollow-imports --windows-icon-from-ico=C:\Users\Yao\Downloads\vse\design\vse.ico --plugin-enable=tk-inter,multiprocessing --output-dir=C:\Users\Yao\Downloads\out --onefile .\gui.py
265274
```
266275

267-
268-
269276
## 社区支持
270277

271278
#### Jetbrains 全家桶支持
@@ -274,3 +281,12 @@ conda install Shapely
274281
<a href="https://jb.gg/OpenSourceSupport"><img src="https://resources.jetbrains.com/storage/products/company/brand/logos/jb_beam.png" alt="JetBrains Logo (Main) logo." width="80"></a>
275282
</div>
276283

284+
## 赞助
285+
<img src="https://i.imgur.com/THE81jM.jpg" width="300">
286+
287+
| 捐赠者 | 累计捐赠金额 | 赞助席位 |
288+
| --- | --- | --- |
289+
| 爱东| 100.00 RMB | 金牌赞助席位 |
290+
| [neoyxm](https://github.com/neoyxm) | 50.00 RMB | 银牌赞助席位 |
291+
| [AcelXiao](https://github.com/acelxiao) | 20.00 RMB | 银牌赞助席位 |
292+
| sky | 5.00 RMB | 铜牌赞助席位 |

README_en.md

+9-7
Original file line numberDiff line numberDiff line change
@@ -12,22 +12,24 @@
1212
- Detect subtitle areas (i.e., coordinates) (as well as scene text if you want) (using deep learning algorithms)
1313
- Converting graphic text into plain-text (using deep learning algorithms)
1414
- Filter non-subtitle text (e.g., Logo and watermark etc.)
15+
- Remove watermark, logo text and original video hard subtitles, see: [video-subtitle-remover (VSR)](https://github.com/YaoFANGUK/video-subtitle-remover/tree/main).
1516
- Remove duplicated subtitle line and **generate srt file** (by calculating text similarity)
1617
- Batch extraction. You can select multiple video files at one time and this tool can generate subtitles for each video.
17-
- Multiple language support, you can extract the following subtitle languages: **Simplified Chinese**, **English**,
18+
- Multiple language support. You can extract subtitles in 87 languages such as: **Simplified Chinese**, **English**,
1819
**Japanese**, **Korean**, **Arabic**, **Traditional Chinese**, **French**, **German**, **Russian**, **Spanish**,
1920
**Portuguese**, **Italian**
2021
- Multiple mode:
21-
- **fast**: high extraction speed while few subtitle missing (**Recommended**)
22-
- **accurate**: no subtitle missing while low extraction speed
22+
- **fast**: use a lightweight model to extract subtitles quickly, but you may lose subtitles (**Recommended**).
23+
- **auto**: automatically judge the model, and use the lightweight model under CPU; Using accurate model under GPU, the speed of subtitle extraction is slow and almost no subtitles are lost.
24+
- **accurate**: use accurate model, detect subtitles frame by frame under GPU, and do not lose subtitles, but the speed is very slow (not recommended).
2325

2426
**Download**
2527

26-
- Windows executable (might be slow when initial start): <a href="https://github.com/YaoFANGUK/video-subtitle-extractor/releases/download/1.0.0/vse.exe">vse.exe</a>
28+
- Windows executable (might be slow when initial start): <a href="https://github.com/YaoFANGUK/video-subtitle-extractor/releases/download/2.0.0/vse.exe">vse.exe</a>
2729

28-
- Windows GPU version:<a href="https://github.com/YaoFANGUK/video-subtitle-extractor/releases/download/1.0.0/vse_windows_GPU.7z">vse_windows_GPU.7z</a>
30+
- Windows GPU version:<a href="https://github.com/YaoFANGUK/video-subtitle-extractor/releases/download/2.0.0/vse_windows_gpu_v2.0.0.7z">vse_windows_gpu_v2.0.0.7z</a>
2931

30-
- Windows CPU version:<a href="https://github.com/YaoFANGUK/video-subtitle-extractor/releases/download/1.0.0/vse_windows_CPU.zip">vse_windows_CPU.zip</a>
32+
- Windows CPU version:<a href="https://github.com/YaoFANGUK/video-subtitle-extractor/releases/download/2.0.0/vse_windows_cpu_v2.0.0.zip">vse_windows_cpu_v2.0.0.zip</a>
3133

3234
- MacOS:<a href="https://github.com/YaoFANGUK/video-subtitle-extractor/releases/download/0.1.0/vse_macOS_CPU.dmg">vse_macOS_CPU.dmg</a>
3335

@@ -38,7 +40,7 @@
3840
- For Command Line Interface(CLI) version, you can escape the subtitle area setting. This program will automatically detect the subtitle area for you.
3941
- GPU support is available. You can install CUDA and cuDNN to speed up the detection and recognition process and even get more accurate results.
4042

41-
<img src="https://z3.ax1x.com/2021/04/09/cNrA1A.png" alt="demo">
43+
<p style="text-align:center;"><img src="https://github.com/YaoFANGUK/video-subtitle-extractor/raw/main/design/demo.png" alt="demo.png"/></p>
4244

4345
> **Provide your suggestions to improve this project in ISSUES**
4446

backend/config.py

+31-35
Original file line numberDiff line numberDiff line change
@@ -76,7 +76,6 @@
7676
# ×××××××××××××××××××× [不要改]判断是否使用GPU start ××××××××××××××××××××
7777

7878

79-
8079
# ×××××××××××××××××××× [不要改]读取语言、模型路径、字典路径 start ××××××××××××××××××××
8180
# 设置识别语言
8281
REC_CHAR_TYPE = settings_config['DEFAULT']['Language']
@@ -88,16 +87,21 @@
8887
ACCURATE_MODE_ON = True
8988
if MODE_TYPE == 'fast':
9089
ACCURATE_MODE_ON = False
90+
if MODE_TYPE == 'auto':
91+
if USE_GPU:
92+
ACCURATE_MODE_ON = True
93+
else:
94+
ACCURATE_MODE_ON = False
9195
# 模型文件目录
92-
# 默认模型版本 V3
93-
MODEL_VERSION = 'V3'
96+
# 默认模型版本 V4
97+
MODEL_VERSION = 'V4'
9498
# 文本检测模型
9599
DET_MODEL_BASE = os.path.join(BASE_DIR, 'models')
96100
# 设置文本识别模型 + 字典
97101
REC_MODEL_BASE = os.path.join(BASE_DIR, 'models')
98102
# 默认字典路径为中文
99103
DICT_BASE = os.path.join(BASE_DIR, 'ppocr', 'utils', 'dict')
100-
# V3模型默认图形识别的shape为3, 48, 320
104+
# V3, V4模型默认图形识别的shape为3, 48, 320
101105
REC_IMAGE_SHAPE = '3,48,320'
102106
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'{REC_CHAR_TYPE}_rec')
103107
DET_MODEL_PATH = os.path.join(DET_MODEL_BASE, MODEL_VERSION, f'{REC_CHAR_TYPE}_det')
@@ -126,58 +130,49 @@
126130

127131
# 定义字典路径
128132
DICT_PATH = os.path.join(DICT_BASE, f'{REC_CHAR_TYPE}_dict.txt')
129-
133+
DET_MODEL_FAST_PATH = os.path.join(DET_MODEL_BASE, 'V4', 'ch_det_fast')
130134
# 如果设置了识别文本语言类型,则设置为对应的语言
131-
if REC_CHAR_TYPE in MULTI_LANG :
132-
# 不管有无GPU和是否开启精准模式,默认使用大模型
135+
if REC_CHAR_TYPE in MULTI_LANG:
133136
# 定义文本识别模型
137+
# 不管有无GPU和是否开启精准模式,默认使用V4版本的大模型
134138
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'{REC_CHAR_TYPE}_rec')
135-
# 如果当前模型版本没有大模型,则切换为V2版本
139+
# 如果默认版本(V4)没有大模型,则切换为默认版本(V4)的fast模型
136140
if not os.path.exists(REC_MODEL_PATH):
137-
MODEL_VERSION = 'V2'
138-
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'{REC_CHAR_TYPE}_rec')
139-
# 如果V2版本也没有大模型版本,则切换V3的fast版本
141+
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'{REC_CHAR_TYPE}_rec_fast')
142+
# 如果默认版本(V4)既没有大模型,又没有fast模型,则使用V3版本的大模型
140143
if not os.path.exists(REC_MODEL_PATH):
141144
MODEL_VERSION = 'V3'
142-
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'{REC_CHAR_TYPE}_rec_fast')
143-
# 如果没有V3的fast版本,则切换V2的fast版本
145+
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'{REC_CHAR_TYPE}_rec')
146+
# 如果V3版本没有大模型,则使用V3版本的fast模型
144147
if not os.path.exists(REC_MODEL_PATH):
145-
MODEL_VERSION = 'V2'
148+
MODEL_VERSION = 'V3'
146149
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'{REC_CHAR_TYPE}_rec_fast')
147150

148151
if REC_CHAR_TYPE in LATIN_LANG:
149-
MODEL_VERSION = 'V3'
150152
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'latin_rec_fast')
151153
DICT_PATH = os.path.join(DICT_BASE, f'latin_dict.txt')
152154
elif REC_CHAR_TYPE in ARABIC_LANG:
153-
MODEL_VERSION = 'V3'
154155
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'arabic_rec_fast')
155156
DICT_PATH = os.path.join(DICT_BASE, f'arabic_dict.txt')
156157
elif REC_CHAR_TYPE in CYRILLIC_LANG:
157-
MODEL_VERSION = 'V3'
158158
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'cyrillic_rec_fast')
159159
DICT_PATH = os.path.join(DICT_BASE, f'cyrillic_dict.txt')
160160
elif REC_CHAR_TYPE in DEVANAGARI_LANG:
161-
MODEL_VERSION = 'V3'
162161
REC_MODEL_PATH = os.path.join(REC_MODEL_BASE, MODEL_VERSION, f'devanagari_rec_fast')
163162
DICT_PATH = os.path.join(DICT_BASE, f'devanagari_dict.txt')
164163

165164
# 定义文本检测模型
166-
tmp_dir = REC_MODEL_PATH.replace(os.path.dirname(REC_MODEL_PATH), '').split('_')
167-
if len(tmp_dir) > 3:
168-
tmp_dir = tmp_dir[-3:]
169-
tmp_dir[1] = 'det'
170-
DET_MODEL_PATH = os.path.join(os.path.dirname(REC_MODEL_PATH), "_".join(tmp_dir))
171-
if not os.path.exists(DET_MODEL_PATH):
172-
if REC_CHAR_TYPE in ['ch', 'japan', 'korean', 'chinese_cht']:
173-
tmp_dir[0] = 'ch'
174-
elif REC_CHAR_TYPE == 'en':
175-
tmp_dir[0] = 'en'
165+
if MODE_TYPE == 'auto':
166+
# 如果使用GPU,则快速模型使用大模型
167+
if USE_GPU:
168+
DET_MODEL_PATH = os.path.join(DET_MODEL_BASE, 'V4', 'ch_det')
176169
else:
177-
tmp_dir[0] = 'ml'
178-
DET_MODEL_PATH = os.path.join(os.path.dirname(REC_MODEL_PATH), "_".join(tmp_dir))
179-
if MODEL_VERSION == 'V2':
180-
DET_MODEL_PATH = os.path.join(DET_MODEL_BASE, MODEL_VERSION, 'ch_det')
170+
DET_MODEL_PATH = os.path.join(DET_MODEL_BASE, 'V4', 'ch_det_fast')
171+
elif MODE_TYPE == 'fast':
172+
DET_MODEL_PATH = os.path.join(DET_MODEL_BASE, 'V4', 'ch_det_fast')
173+
else:
174+
DET_MODEL_PATH = os.path.join(DET_MODEL_BASE, 'V4', 'ch_det')
175+
181176
# 定义图像识别shape
182177
if MODEL_VERSION == 'V2':
183178
REC_IMAGE_SHAPE = '3,32,320'
@@ -188,6 +183,10 @@
188183
if 'inference.pdiparams' not in (os.listdir(REC_MODEL_PATH)):
189184
fs = Filesplit()
190185
fs.merge(input_dir=REC_MODEL_PATH)
186+
# 查看该路径下是否有文本模型识别完整文件,没有的话合并小文件生成完整文件
187+
if 'inference.pdiparams' not in (os.listdir(DET_MODEL_PATH)):
188+
fs = Filesplit()
189+
fs.merge(input_dir=DET_MODEL_PATH)
191190
# ×××××××××××××××××××× [不要改]读取语言、模型路径、字典路径 end ××××××××××××××××××××
192191

193192

@@ -215,9 +214,6 @@
215214
# 每几帧抽取一帧进行OCR识别
216215
EXTRACT_INTERVAL = 8
217216

218-
# 欧式距离相似值
219-
EUCLIDEAN_SIMILARITY_THRESHOLD = 0.9
220-
221217
# 容忍的像素点偏差
222218
PIXEL_TOLERANCE_Y = 50 # 允许检测框纵向偏差50个像素点
223219
PIXEL_TOLERANCE_X = 100 # 允许检测框横向偏差100个像素点

backend/interface/ch.ini

+2-1
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@ Title = 字幕提取器
33
InterfaceLanguage = 选择语言:
44
SubtitleLanguage = 选择视频字幕的语言:
55
Mode = 选择识别模式:
6+
ModeAuto = 自动
67
ModeFast = 快速
78
ModeAccurate = 精准
89
InterfaceDefault = 简体中文
@@ -135,4 +136,4 @@ CheckSubArea = 请查看图片, 确定字幕区域是否正确:
135136
DeleteNoSubArea = 红色框区域外的字幕是否去除? 输入 "y""回车" 表示去除,输入"n"或其他表示不去除:
136137
FinishDeleteNoSubArea = 去除完毕
137138
SubLocation = 字幕文件生成位置:
138-
InputVideo = 请输入视频完整路径:
139+
InputVideo = 请输入视频完整路径:

backend/interface/chinese_cht.ini

+2-1
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@ Title = 字幕提取器
33
InterfaceLanguage = 選擇語言:
44
SubtitleLanguage = 選擇視頻字幕的語言:
55
Mode = 選擇識別模式:
6+
ModeAuto = 自動
67
ModeFast = 快速
78
ModeAccurate = 精準
89
InterfaceDefault = 繁體中文
@@ -135,4 +136,4 @@ CheckSubArea = 請查看圖片, 確定字幕區域是否正確:
135136
DeleteNoSubArea = 紅色框區域外的字幕是否去除? 輸入 "y""回車" 表示去除,輸入"n"或其他表示不去除:
136137
FinishDeleteNoSubArea = 去除完畢
137138
SubLocation = 字幕文件生成位置:
138-
InputVideo = 請輸入視頻完整路徑:
139+
InputVideo = 請輸入視頻完整路徑:

backend/interface/en.ini

+2-1
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@ Title = Subtitle Extractor
33
InterfaceLanguage = Choose Language:
44
SubtitleLanguage = Subtitle Language:
55
Mode = Choose Mode:
6+
ModeAuto = auto
67
ModeFast = fast
78
ModeAccurate = accurate
89
InterfaceDefault = English
@@ -135,4 +136,4 @@ CheckSubArea = Please check the picture to make sure the subtitle area is correc
135136
DeleteNoSubArea = Are the subtitles outside the red box area removed? Input "y" or "Enter" to remove, input "n" or other means not to remove:
136137
FinishDeleteNoSubArea = Removed
137138
SubLocation = Subtitle file generated at:
138-
InputVideo = Please enter the full path of the video:
139+
InputVideo = Please enter the full path of the video:

backend/interface/ko.ini

+2-1
Original file line numberDiff line numberDiff line change
@@ -3,6 +3,7 @@ Title = Subtitle Extractor
33
InterfaceLanguage = 언어 선택:
44
SubtitleLanguage = 자막 언어:
55
Mode = 모드 선택:
6+
ModeAuto = 자동적 인
67
ModeFast = 빠름
78
ModeAccurate = 정확함
89
InterfaceDefault = 한국어
@@ -135,4 +136,4 @@ CheckSubArea = 자막 영역이 올바른지 이미지를 확인해주세요.
135136
DeleteNoSubArea = 빨간색 상자 영역 밖의 자막이 제거되었나요? 제거하려면 "y" 또는 "Enter"를 입력하고, 제거하지 않으려면 "n" 또는 아무거나 입력합니다:
136137
FinishDeleteNoSubArea = 제거 완료
137138
SubLocation = 자막 파일이 생성된 위치:
138-
InputVideo = 비디오의 전체 경로를 입력하세요:
139+
InputVideo = 비디오의 전체 경로를 입력하세요:

0 commit comments

Comments
 (0)