Update audio.py

audio.py

@@ -5,11 +5,9 @@ import tempfile
 import os
 
 os.environ["IMAGEIO_FFMPEG_EXE"] = "/usr/local/bin/ffmpeg"
-# 如果需要，设置 ImageMagick 的路径
 # os.environ["IMAGEMAGICK_BINARY"] = "/usr/local/bin/convert"
 
 def gen_audio(subtitles, tmpdir):
-# 字幕列表，表示在指定时间段显示对应的字幕
 # subtitles = [
 #    [[0, 3], "這裡請注意右臂"],
 #     [[4, 8], "這裡請注意左臂"],
@@ -20,45 +18,37 @@ def gen_audio(subtitles, tmpdir):
     for subtitle in subtitles:
         print(subtitle)
     combined_clip = VideoFileClip(tmpdir + "/output.mp4")
-    # 获取视频总时长
     total_duration = combined_clip.duration
 
-    # 添加字幕的函数
     def add_subtitles(clip, subtitles):
         txt_clips = []
         for start_end, text in subtitles:
             start_time, end_time = start_end
             duration = end_time - start_time
-            txt_clip = (TextClip(text, fontsize=clip.w//20, color='white', font='bold.ttf', method='caption', size=(clip.w * 0.9, None))  # 添加 method='caption'
-                        .set_duration(duration)  # 设置字幕持续时间
+            txt_clip = (TextClip(text, fontsize=clip.w//20, color='white', font='bold.ttf', method='caption', size=(clip.w * 0.9, None))  
+                        .set_duration(duration)  
                         .set_position(('center', 4 * clip.h // 5))
-                        .set_start(start_time))  # 字幕开始时间
+                        .set_start(start_time))
             txt_clips.append(txt_clip)
         
-        # 将字幕叠加到视频上
         return CompositeVideoClip([clip, *txt_clips])
 
-    # 将字幕添加到视频中
     video_with_subtitles = add_subtitles(combined_clip, subtitles)
 
-    # 生成语音文件并返回音频剪辑的函数
     def generate_audio(subtitles, total_duration):
-        temp_audio_path = tempfile.mkdtemp()  # 创建临时目录存储音频文件
+        temp_audio_path = tempfile.mkdtemp()  
         clips = []
 
         for i, (start_end, text) in enumerate(subtitles):
             start_time, end_time = start_end
             duration = end_time - start_time
 
-            # 使用 gtts 生成语音
             tts = gTTS(text=text, lang='zh')
             audio_path = os.path.join(temp_audio_path, f'subtitle_{i}.mp3')
             tts.save(audio_path)
             
-            # 使用 pydub 加载音频并将其转换为 wav 格式
             audio_segment = AudioSegment.from_mp3(audio_path)
 
-            # 将 pydub 音频转换为 moviepy 可用的 AudioFileClip
             new_audio_path = os.path.join(temp_audio_path, f'subtitle_{i}_adjusted.wav')
             audio_segment.export(new_audio_path, format="wav")
 
@@ -66,17 +56,13 @@ def gen_audio(subtitles, tmpdir):
             
             clips.append(audio_clip)
         
-        # 合并所有音频剪辑
         final_audio = CompositeAudioClip(clips)
         return final_audio
 
-    # 生成语音
     audio_clip = generate_audio(subtitles, total_duration)
 
-    # 将语音添加到视频
     video_with_audio = video_with_subtitles.set_audio(audio_clip)
 
-    # 导出带语音和字幕的视频
     video_with_audio.write_videofile(tmpdir + '/output_with_subtitles.mp4', fps=24)
     return tmpdir + '/output_with_subtitles.mp4'
 
@@ -87,35 +73,29 @@ def merge_subtitles(subtitles):
     n = len(subtitles)
     
     while i < n:
-        # 初始化当前字幕的起始时间、结束时间和内容
         current_start, current_end = subtitles[i][0]
         current_content = subtitles[i][1]
         duration = current_end - current_start
-        i += 1  # 移动到下一个字幕索引，供后续合并使用
+        i += 1  
 
-        # 尝试合并后续字幕，直到持续时间至少为3秒
         while duration < 3:
             if i >= n:
-                # 已经处理完所有字幕，延长结束时间
                 current_end = current_start + 3
                 duration = current_end - current_start
                 break
             next_start, next_end = subtitles[i][0]
             next_content = subtitles[i][1]
             if next_start == current_end:
-                # 与下一个字幕相邻，合并
                 current_end = next_end
                 current_content += "、" + next_content
                 duration = current_end - current_start
-                i += 1  # 继续检查下一个字幕
+                i += 1  
             else:
-                # 不相邻，无法合并，延长结束时间
                 if duration < 3:
                     current_end = current_start + 3
                     duration = current_end - current_start
-                break  # 退出合并循环
+                break  
 
-        # 添加合并后的字幕到新列表
         current_content = '這裡請注意' + '、'.join(list(set(current_content.replace("細節", "").split('、'))))
         merged_subtitles.append([[current_start, current_end], current_content])