Upload 10 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+examples/demo_shejipuhui.mp4 filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,168 @@
+import whisper
+import gradio as gr
+import os
+import datetime
+
+#获取当前北京时间
+utc_dt = datetime.datetime.utcnow()
+beijing_dt = utc_dt.astimezone(datetime.timezone(datetime.timedelta(hours=16)))
+formatted = beijing_dt.strftime("%Y-%m-%d_%H")
+print(f"北京时间: {beijing_dt.year}年{beijing_dt.month}月{beijing_dt.day}日 "
+      f"{beijing_dt.hour}时{beijing_dt.minute}分{beijing_dt.second}秒")
+#创建作品存放目录
+works_path = '../works_audio_video_transcribe/' + formatted
+if not os.path.exists(works_path):
+  os.makedirs(works_path)
+print('作品目录：' + works_path)
+
+modelname = "small"
+model = whisper.load_model(modelname) #tiny、base、small、medium（可用）、large
+
+def transcript(audiofile, prompt, output_dir):
+    os.system(f"whisper {audiofile} --model {modelname} --language zh --initial_prompt {prompt} --output_dir {output_dir}")
+
+def audio_recog(audiofile):
+    utc_dt = datetime.datetime.utcnow()
+    beijing_dt = utc_dt.astimezone(datetime.timezone(datetime.timedelta(hours=16)))
+    formatted = beijing_dt.strftime("%Y-%m-%d_%H-%M-%S")
+    print(f"开始时间: {beijing_dt.year}年{beijing_dt.month}月{beijing_dt.day}日 "
+      f"{beijing_dt.hour}时{beijing_dt.minute}分{beijing_dt.second}秒")
+      
+    print("音频文件：" + audiofile)
+    
+    prompt = "以下是普通话的句子"
+    filename = os.path.splitext(os.path.basename(audiofile))[0]
+    output_txt = works_path + '/' + filename + '.txt'
+    output_srt = works_path + '/' + filename + '.srt'
+    
+    output_dir = works_path
+    transcript(audiofile, prompt, output_dir)
+    with open(output_txt, "r") as f:
+        text_output = f.read()
+        print("text：" + text_output)
+    print("text文件：" + output_txt)
+
+    with open(output_srt, "r") as f:
+        srt_output = f.read()
+        print("srt：" + srt_output)
+    print("srt文件：" + output_srt)
+    
+    utc_dt = datetime.datetime.utcnow()
+    beijing_dt = utc_dt.astimezone(datetime.timezone(datetime.timedelta(hours=16)))
+    formatted = beijing_dt.strftime("%Y-%m-%d_%H-%M-%S")
+    print(f"结束时间: {beijing_dt.year}年{beijing_dt.month}月{beijing_dt.day}日 "
+      f"{beijing_dt.hour}时{beijing_dt.minute}分{beijing_dt.second}秒")
+    
+    return text_output, srt_output
+
+def video_recog(filepath):
+    utc_dt = datetime.datetime.utcnow()
+    beijing_dt = utc_dt.astimezone(datetime.timezone(datetime.timedelta(hours=16)))
+    formatted = beijing_dt.strftime("%Y-%m-%d_%H-%M-%S")
+    print(f"开始时间: {beijing_dt.year}年{beijing_dt.month}月{beijing_dt.day}日 "
+      f"{beijing_dt.hour}时{beijing_dt.minute}分{beijing_dt.second}秒")
+      
+    filename = os.path.splitext(os.path.basename(filepath))[0]
+    worksfile = works_path + '/works_' + filename + '.mp4'
+    print("视频文件：" + filepath)
+    
+    # 提取音频为aac
+    audiofile = works_path + '/' + filename + '.mp3'
+    os.system(f"ffmpeg -i {filepath} -vn -c:a libmp3lame -q:a 4 {audiofile}")
+    print("音频文件：" + audiofile)
+    
+    prompt = "以下是普通话的句子"
+    output_txt = works_path + '/' + filename + '.txt'
+    output_srt = works_path + '/' + filename + '.srt'
+    
+    output_dir = works_path
+    transcript(audiofile, prompt, output_dir)
+    with open(output_txt, "r") as f:
+        text_output = f.read()
+        print("text：" + text_output)
+    print("text文件：" + output_txt)
+
+    with open(output_srt, "r") as f:
+        srt_output = f.read()
+        print("srt：" + srt_output)
+    print("srt文件：" + output_srt)
+    
+    # 给视频添加字幕
+    os.system(f"ffmpeg -i {filepath} -i {output_srt} -c:s mov_text -c:v copy -c:a copy {worksfile}")
+    print("作品：" + worksfile)
+    
+    utc_dt = datetime.datetime.utcnow()
+    beijing_dt = utc_dt.astimezone(datetime.timezone(datetime.timedelta(hours=16)))
+    formatted = beijing_dt.strftime("%Y-%m-%d_%H-%M-%S")
+    print(f"结束时间: {beijing_dt.year}年{beijing_dt.month}月{beijing_dt.day}日 "
+      f"{beijing_dt.hour}时{beijing_dt.minute}分{beijing_dt.second}秒")
+    
+    return text_output, srt_output, worksfile
+
+css_style = "#fixed_size_img {height: 240px;} " \
+            "#overview {margin: auto;max-width: 400px; max-height: 400px;}"
+
+title = "音视频转录 by宁侠"
+description = "您只需要输入一张物体照片和背景图，即可快速完成背景替换，让物体仿佛置身于不同的场景中，实现穿越体验！无论是漫游古代的宫殿，还是探索未来的科技世界，我们的系统能让您轻松达成，让您的想象力得以自由驰骋！快来体验吧，欢迎使用我们的通用换景服务！"
+
+examples_path = 'examples/'
+examples = [[examples_path + 'demo_shejipuhui.mp4']]
+
+# gradio interface
+with gr.Blocks(title=title, css=css_style) as demo:
+    gr.HTML('''
+      <div style="text-align: center; max-width: 720px; margin: 0 auto;">
+                  <div
+                    style="
+                      display: inline-flex;
+                      align-items: center;
+                      gap: 0.8rem;
+                      font-size: 1.75rem;
+                    "
+                  >
+                    <h1 style="font-family:  PingFangSC; font-weight: 500; font-size: 36px; margin-bottom: 7px;">
+                      音视频转录
+                    </h1>
+                    <h1 style="font-family: PingFangSC; font-weight: 500; line-height: 1.5em; font-size: 16px; margin-bottom: 7px;">
+                      by宁侠
+                    </h1>
+      ''')
+    gr.Markdown(description)
+    video_state = gr.State()
+    audio_state = gr.State()
+    with gr.Tab("🎥视频转录 Video Transcribe"):
+        with gr.Row():
+            with gr.Column():
+                video_input = gr.Video(label="🎥视频输入 Video Input")
+                gr.Examples(['examples/demo_shejipuhui.mp4'], [video_input], label='语音识别示例 ASR Demo')
+                video_recog_button = gr.Button("👂视频识别 Recognize")
+                video_output = gr.Video(label="🎥视频 Video", visible=False)
+            with gr.Column():
+                video_text_output = gr.Textbox(label="✏️识别结果 Recognition Result")
+                video_srt_output = gr.Textbox(label="📖SRT字幕内容 SRT Subtitles")
+                with gr.Row(visible=False):
+                    font_size = gr.Slider(minimum=10, maximum=100, value=32, step=2, label="🔠字幕字体大小 Subtitle Font Size")
+                    font_color = gr.Radio(["black", "white", "green", "red"], label="🌈字幕颜色 Subtitle Color", value='white')
+                video_subtitles_button = gr.Button("添加字幕\nGenerate Subtitles", visible=False)
+                
+    
+    video_recog_button.click(video_recog, inputs=[video_input], outputs=[video_text_output, video_srt_output, video_output])
+#    video_subtitles_button.click(video_subtitles, inputs=[video_text_input], outputs=[video_output])
+
+    with gr.Tab("🔊音频转录 Audio Transcribe"):
+        with gr.Row():
+            with gr.Column():
+                audio_input = gr.Audio(label="🔊音频输入 Audio Input", type="filepath")
+                gr.Examples(['examples/paddlespeech.asr-zh.wav', 'examples/demo_shejipuhui.mp3'], [audio_input])
+                audio_recog_button = gr.Button("👂音频识别 Recognize")
+            with gr.Column():
+                audio_text_output = gr.Textbox(label="✏️识别结果 Recognition Result")
+                audio_srt_output = gr.Textbox(label="📖SRT字幕内容 SRT Subtitles")
+                audio_subtitles_button = gr.Button("添加字幕\nGenerate Subtitles", visible=False)
+                audio_output = gr.Audio(label="🔊音频 Audio", visible=False)
+    
+    audio_recog_button.click(audio_recog, inputs=[audio_input], outputs=[audio_text_output, audio_srt_output])
+#    audio_subtitles_button.click(audio_subtitles, inputs=[audio_text_input], outputs=[audio_output])
+
+# start gradio service in local
+demo.queue(api_open=False).launch(debug=True)

examples/add_subtitles.py

@@ -0,0 +1,28 @@
+#from moviepy.editor import *
+#from moviepy.video.VideoClip import VideoClip
+#from moviepy.video.tools.subtitles import SubtitlesClip
+#
+### 或者直接使用TextClip
+##subtitle = TextClip("Hello World", fontsize=24, color='white')
+#
+#video = VideoFileClip("demo_shejipuhui.mp4")
+##subtitles = (SubtitlesClip("demo_shejipuhui.srt", font="Arial", fontsize=24, color='white'))
+#subtitles = (SubtitlesClip("demo_shejipuhui.srt"))
+#
+#final = CompositeVideoClip([video, subtitles])
+#final.write_videofile("output.mp4")
+
+#import ffmpeg
+import subprocess
+
+command = "ffmpeg -i demo_shejipuhui.mp4 -i demo_shejipuhui.srt -c:s mov_text -c:v copy -c:a copy output.mp4"
+subprocess.call(command, shell=True)
+
+#video = ffmpeg.input('demo_shejipuhui.mp4')
+#subtitle = ffmpeg.input('demo_shejipuhui.srt')
+#
+#out = ffmpeg.concat(video, subtitle, v=1, a=1).node
+#out[0].opts = '-codec copy'
+#out[1].opts = '-codec mov_text'
+#
+#out.output('output.mp4').run()

examples/demo_shejipuhui.ass

@@ -0,0 +1,27 @@
+[Script Info]
+; Script generated by FFmpeg/Lavc58.54.100
+ScriptType: v4.00+
+PlayResX: 384
+PlayResY: 288
+
+[V4+ Styles]
+Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
+Style: Default,Arial,16,&Hffffff,&Hffffff,&H0,&H0,0,0,0,0,100,100,0,0,1,1,0,2,10,10,10,0
+
+[Events]
+Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
+Dialogue: 0,0:00:00.00,0:00:01.44,Default,,0,0,0,,在我们的设计普惠当中
+Dialogue: 0,0:00:01.44,0:00:03.76,Default,,0,0,0,,有一个我经常津津乐道的项目
+Dialogue: 0,0:00:03.76,0:00:06.08,Default,,0,0,0,,叫寻找远方的美好
+Dialogue: 0,0:00:07.08,0:00:09.08,Default,,0,0,0,,在这样一个 我们叫寻美
+Dialogue: 0,0:00:09.08,0:00:10.64,Default,,0,0,0,,在这样的一个项目当中
+Dialogue: 0,0:00:10.64,0:00:13.24,Default,,0,0,0,,我们把它跟乡村振兴去结合起来
+Dialogue: 0,0:00:13.24,0:00:15.28,Default,,0,0,0,,利用我们的设计的能力
+Dialogue: 0,0:00:15.28,0:00:17.56,Default,,0,0,0,,我们自身员工的设计能力
+Dialogue: 0,0:00:17.56,0:00:19.60,Default,,0,0,0,,我们设计生态伙伴的能力
+Dialogue: 0,0:00:19.60,0:00:22.60,Default,,0,0,0,,帮助乡村振兴当中
+Dialogue: 0,0:00:22.60,0:00:25.64,Default,,0,0,0,,要希望把他的产品推向市场
+Dialogue: 0,0:00:25.64,0:00:27.44,Default,,0,0,0,,把他的农产品 把他的加工产品
+Dialogue: 0,0:00:27.48,0:00:30.04,Default,,0,0,0,,推向市场的这样的伙伴
+Dialogue: 0,0:00:30.04,0:00:31.16,Default,,0,0,0,,做一件事情
+Dialogue: 0,0:00:31.16,0:00:33.76,Default,,0,0,0,,就是帮他们设计一个好的包装物

examples/demo_shejipuhui.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8fa2612b7a25e94f8ec3fe96ac88fb01874b8bbed5b7bc10d07ef0555340bc6
+size 4784476

examples/demo_shejipuhui.srt

@@ -0,0 +1,60 @@
+1
+00:00:00,000 --> 00:00:01,440
+在我们的设计普惠当中
+
+2
+00:00:01,440 --> 00:00:03,760
+有一个我经常津津乐道的项目
+
+3
+00:00:03,760 --> 00:00:06,080
+叫寻找远方的美好
+
+4
+00:00:07,080 --> 00:00:09,080
+在这样一个 我们叫寻美
+
+5
+00:00:09,080 --> 00:00:10,640
+在这样的一个项目当中
+
+6
+00:00:10,640 --> 00:00:13,240
+我们把它跟乡村振兴去结合起来
+
+7
+00:00:13,240 --> 00:00:15,280
+利用我们的设计的能力
+
+8
+00:00:15,280 --> 00:00:17,560
+我们自身员工的设计能力
+
+9
+00:00:17,560 --> 00:00:19,600
+我们设计生态伙伴的能力
+
+10
+00:00:19,600 --> 00:00:22,600
+帮助乡村振兴当中
+
+11
+00:00:22,600 --> 00:00:25,640
+要希望把他的产品推向市场
+
+12
+00:00:25,640 --> 00:00:27,440
+把他的农产品 把他的加工产品
+
+13
+00:00:27,480 --> 00:00:30,040
+推向市场的这样的伙伴
+
+14
+00:00:30,040 --> 00:00:31,160
+做一件事情
+
+15
+00:00:31,160 --> 00:00:33,760
+就是帮他们设计一个好的包装物
+

examples/demo_shejipuhui.txt

@@ -0,0 +1,15 @@
+在我们的设计普惠当中
+有一个我经常津津乐道的项目
+叫寻找远方的美好
+在这样一个 我们叫寻美
+在这样的一个项目当中
+我们把它跟乡村振兴去结合起来
+利用我们的设计的能力
+我们自身员工的设计能力
+我们设计生态伙伴的能力
+帮助乡村振兴当中
+要希望把他的产品推向市场
+把他的农产品 把他的加工产品
+推向市场的这样的伙伴
+做一件事情
+就是帮他们设计一个好的包装物

requirements.txt

@@ -0,0 +1,2 @@
+openai-whisper
+git+https://github.com/openai/whisper.git