1.[modified]将界面大小固定，并对各组件均设置为左对齐;

.gitignore

+/dist/
+/build/
+/missing_packages/

README.md

+### 文件结构
+
+1、界面文件——try_with_gui.py
+
+2、各功能模块：
+
+- judge_subtitle.py：检测视频是否提供字幕
+- detect_with_ocr.py：在有字幕前提下使用ocr获取视频中的字幕
+- detect_with_asr.py：在无字幕前提下使用asr获取视频中的字幕
+- narratage_detection.py：获取视频字幕及对应时间戳
+- speech_synthesis.py：生成旁白语音
+- split_wav.py：对视频中的音频进行提取、切分等操作
+- PaddlePaddle_DeepSpeech2文件夹：使用PaddleSpeech获取音频对应文本
+

speech_synthesis.py

@@ -49,7 +49,6 @@ def change_speed(wav_path, speed=1.0):
     :param speed: 转换后的语速
     :return:
     """
-    print("调速")
     cmd_line = 'ffmpeg -y -i {} -filter:a \"atempo={}\" {}'.format(tmp_file, speed, wav_path)
     os.system(cmd_line)
 
@@ -92,7 +91,6 @@ def get_narratage_text(sheet_content, speed):
     narratage_end_time = []
     narratage_text = []
     for i, text in enumerate(narratage):
-        print(i, text)
         if text is not None:
             if text == '翻译':
                 narratage_text.append(subtitle[i])
@@ -161,7 +159,6 @@ def adjust_volume(origin, start_timestamp, end_timestamp):
 
 def mix_speech(origin, narratage_paths, start_timestamps):
     composed_wav_path = os.path.join(os.path.dirname(origin), "composed.wav")
-    print(composed_wav_path)
     command_line = 'ffmpeg -i {}'.format(origin)
     for i, narratage_path in enumerate(narratage_paths):
         command_line += " -i {}".format(narratage_path)
@@ -172,7 +169,6 @@ def mix_speech(origin, narratage_paths, start_timestamps):
     command_line = command_line + "".join(["[aud{}]".format(str(i + 1)) for i in range(len(start_timestamps))])
     command_line += "amix=inputs={}\" -vsync 2 -y {}".format(len(start_timestamps) + 1, composed_wav_path)
     os.system(command_line)
-    print(command_line)
 
 
 def ss_and_export(video_path, sheet_path, output_dir, speed, caption_file, state=None):
@@ -244,12 +240,13 @@ if __name__ == '__main__':
     # video_path, sheet_path, output_dir, speed, caption_file = args.video_path,\
     # args.sheet_path, args.output_dir, args.speed, args.caption_file
 
-    video_path = 'D:/heelo/hysxm_3.mp4'
-    sheet_path = 'D:/heelo/hysxm_3.xlsx'
-    output_dir = 'D:/AddCaption/hysxm_3'
-    speed = 1.25
-    caption_file = 'D:/AddCaption/hysxm_3/hysxm_3.srt'
-
-    # 主函数执行
-    ss_and_export(video_path=video_path, sheet_path=sheet_path, output_dir=output_dir, speed=speed,
-                  caption_file=caption_file)
+    # video_path = 'D:/heelo/hysxm_3.mp4'
+    # sheet_path = 'D:/heelo/hysxm_3.xlsx'
+    # output_dir = 'D:/AddCaption/hysxm_3'
+    # speed = 1.25
+    # caption_file = 'D:/AddCaption/hysxm_3/hysxm_3.srt'
+    #
+    # # 主函数执行
+    # ss_and_export(video_path=video_path, sheet_path=sheet_path, output_dir=output_dir, speed=speed,
+    #               caption_file=caption_file)
+    pass
\ No newline at end of file

split_wav.py

@@ -89,4 +89,4 @@ if __name__ == '__main__':
     video_path = 'D:/heelo/shaolin.mkv'
     extract_audio(video_path, './tmp', 62, 7489)
     split_audio()
-    extrac_speech()
+    extract_speech()

try_with_gui.py

@@ -11,6 +11,7 @@ import inspect
 window = tk.Tk()
 window.title('无障碍电影辅助工具')  # 标题
 window.geometry('600x400')  # 窗口尺寸
+window.resizable(0, 0)
 
 
 def open_video_file():
@@ -75,6 +76,7 @@ def start_process(p, p_label, state, intervals=100):
     启动进度条
     :param p: 进度条组件
     :param p_label: 进度条对应百分比文本
+    :param state: 进度条与任务用于通信对齐的变量，代表任务的实际进度
     :param intervals: 进度条前进所需时间
     :return:
     """
@@ -86,7 +88,7 @@ def start_process(p, p_label, state, intervals=100):
         if state[0] and state[0] != lastState and state[0] * 100 > p['value']:
             p['value'] = round(state[0] * 100, 2)
             lastState = state[0]
-        p_label['text'] = str(round(p['value'], 2)) + "%"
+        p_label['text'] = format(format(p['value'], '2.2f'), "5s") + "%"
         if p['value'] == 100.0:
             p.stop()
             p['value'] = 100.0
@@ -118,9 +120,8 @@ def start_detect():
     from narratage_detection import detect
     # 显示进度条及开始检测
     progressbar_1.grid(column=2, row=1, sticky="W")
-    progress_1.grid(column=3, row=1)
+    progress_1.grid(column=3, row=1, sticky="W")
     processState.set("开始检测")
-    intervals = trans_to_seconds(endTime.get())
     # 多线程同步进行检测和进度条更新
     state = [None]
     threads = [
@@ -135,7 +136,7 @@ def start_detect():
         t.join()
     # 将进度条的进度拉满到100%，并给出“任务已完成”的提示
     progressbar_1['value'] = 100
-    progress_1['text'] = '100%'
+    progress_1['text'] = '100.0%'
     processState.set("任务已完成")
     # 检测完成后，将“停止检测”按钮设置为不可点击状态，”开始检测“按钮设置为可点击状态
     stopDetection.config(state=tk.DISABLED)
@@ -152,7 +153,7 @@ def stop_detect():
     startDetection.config(state=tk.ACTIVE)
     progressbar_1.stop()
     progressbar_1['value'] = 0
-    progress_1['text'] = "0%"
+    progress_1['text'] = "00.00%"
     progressbar_1.grid_forget()
     progress_1.grid_forget()
 
@@ -219,6 +220,8 @@ def start_synthesis():
     elif not os.path.exists(audio_dir):
         messagebox.showwarning("警告", "当前音频存放路径有误，请检查一遍。")
         return
+    if len(caption_path) == 0:
+        messagebox.showwarning("警告", "请选择字幕文件存放路径")
     if len(sheet_path) == 0:
         messagebox.showwarning("警告", "请选择你要处理的表格")
         return
@@ -258,7 +261,7 @@ def stop_synthesis():
     startSynthesis.config(state=tk.ACTIVE)
     progressbar_2.stop()
     progressbar_2['value'] = 0
-    progress_2['text'] = "0%"
+    progress_2['text'] = "00.00%"
     progressbar_2.grid_forget()
     progress_2.grid_forget()
 
@@ -317,13 +320,18 @@ tabControl.pack(expand=1, fill="both")
 video_info = ttk.LabelFrame(tab1, text=" 视频信息操作 ")
 video_info.place(relx=0.05, rely=0.05, relwidth=0.9, relheight=0.4)
 
+rows = 4
+for i in range(rows):
+    video_info.grid_rowconfigure(i, weight=1)
+    video_info.grid_columnconfigure(i, weight=1)
+
 input_label = ttk.Label(video_info, text="视频文件")
 input_label.grid(column=0, row=0)
 inputFilePath = tk.StringVar()
 inputFile = ttk.Entry(video_info, width=30, textvariable=inputFilePath)
-inputFile.grid(column=1, row=0)
+inputFile.grid(column=1, row=0, sticky="W")
 upload_button = ttk.Button(video_info, text="上传文件", command=open_video_file)
-upload_button.grid(column=2, row=0)
+upload_button.grid(column=2, row=0, sticky="W")
 
 startTime_label = ttk.Label(video_info, text="视频实际开始时间")
 startTime_label.grid(column=0, row=1)
@@ -354,15 +362,20 @@ choice_3.grid(column=3, row=3, sticky="W")
     - 停止检测按钮
 """
 detect_command = ttk.LabelFrame(tab1, text=" 检测步骤 ")
-detect_command.place(relx=0.05, rely=0.5, relwidth=0.9, relheight=0.4)
+detect_command.place(relx=0.05, rely=0.6, relwidth=0.9, relheight=0.3)
+
+rows = 4
+for i in range(rows):
+    detect_command.grid_rowconfigure(i, weight=1)
+    detect_command.grid_columnconfigure(i, weight=1)
 
 output_label = ttk.Label(detect_command, text="输出表格")
 output_label.grid(column=0, row=0)
 outputFilePath = tk.StringVar()
 outputFile = ttk.Entry(detect_command, width=30, textvariable=outputFilePath)
-outputFile.grid(column=1, row=0)
+outputFile.grid(column=1, row=0, sticky="W")
 save_button = ttk.Button(detect_command, text="打开文件夹", command=find_save_file)
-save_button.grid(column=2, row=0)
+save_button.grid(column=2, row=0, sticky="W")
 
 startDetection = ttk.Button(detect_command, text="开始检测", command=lambda: thread_it(start_detect, name="startDetect"))
 startDetection.grid(column=0, row=1)
@@ -388,11 +401,16 @@ stopDetection.config(state=tk.DISABLED)
 audio_info = ttk.LabelFrame(tab2, text=" 语音相关设置 ")
 audio_info.place(relx=0.05, rely=0.05, relwidth=0.9, relheight=0.3)
 
+rows = 3
+for i in range(rows):
+    audio_info.grid_rowconfigure(i, weight=1)
+    audio_info.grid_columnconfigure(i, weight=1)
+
 video_label = ttk.Label(audio_info, text="原视频")
 video_label.grid(column=0, row=0)
 videoPath = tk.StringVar()
 videoPath_input = ttk.Entry(audio_info, width=30, textvariable=videoPath)
-videoPath_input.grid(column=1, row=0)
+videoPath_input.grid(column=1, row=0, sticky="W")
 upload_button_3 = ttk.Button(audio_info, text="上传文件", command=confirm_video_path)
 upload_button_3.grid(column=2, row=0)
 
@@ -400,7 +418,7 @@ narratage_label = ttk.Label(audio_info, text="旁白脚本表格")
 narratage_label.grid(column=0, row=1)
 narratagePath = tk.StringVar()
 narratagePath_input = ttk.Entry(audio_info, width=30, textvariable=narratagePath)
-narratagePath_input.grid(column=1, row=1)
+narratagePath_input.grid(column=1, row=1, sticky="W")
 upload_button_2 = ttk.Button(audio_info, text="上传文件", command=open_sheet_file)
 upload_button_2.grid(column=2, row=1)
 
@@ -421,7 +439,12 @@ speedChosen.grid(column=1, row=2, sticky="W")
     - 停止合成按钮
 """
 synthesis_command = ttk.LabelFrame(tab2, text=" 语音合成步骤 ")
-synthesis_command.place(relx=0.05, rely=0.45, relwidth=0.9, relheight=0.5)
+synthesis_command.place(relx=0.05, rely=0.45, relwidth=0.9, relheight=0.4)
+
+rows = 4
+for i in range(rows):
+    synthesis_command.grid_rowconfigure(i, weight=1)
+    synthesis_command.grid_columnconfigure(i, weight=1)
 
 audioDir_label = ttk.Label(synthesis_command, text="输出音频存放于")
 audioDir_label.grid(column=0, row=0)
@@ -446,7 +469,7 @@ processState_2 = tk.StringVar()
 stateLabel_2 = tk.Label(synthesis_command, textvariable=processState_2, fg="green")
 stateLabel_2.grid(column=1, row=2, sticky="W")
 progressbar_2 = ttk.Progressbar(synthesis_command, length=100, mode="determinate")
-progress_2 = tk.Label(synthesis_command, text="0%")
+progress_2 = tk.Label(synthesis_command, text="00.00%")
 stopSynthesis = ttk.Button(synthesis_command, text="停止合成",
                            command=lambda: thread_it(stop_synthesis, name="stopSynthesis"))
 stopSynthesis.grid(column=0, row=3)