Update audio.py

audio.py

@@ -3,6 +3,8 @@ from gtts import gTTS
 from pydub import AudioSegment
 import tempfile
 import os
+import pyttsx3
+from ms_tts import ms_tts_gen
 
 os.environ["IMAGEIO_FFMPEG_EXE"] = "/usr/local/bin/ffmpeg"
 # os.environ["IMAGEMAGICK_BINARY"] = "/usr/local/bin/convert"
@@ -12,7 +14,7 @@ def gen_audio(subtitles, tmpdir):
 #    [[0, 3], "這裡請注意右臂"],
 #     [[4, 8], "這裡請注意左臂"],
 #     [[9, 12], "這裡請注意左手細節"]
-# ]
+# ])
     subtitles = merge_subtitles(subtitles)
     print("===> Subtitles:")
     for subtitle in subtitles:
@@ -44,13 +46,19 @@ def gen_audio(subtitles, tmpdir):
         temp_audio_path = tempfile.mkdtemp()  
         clips = []
 
+        # engine = pyttsx3.init()
+        # engine.setProperty('voice', 'yue') # yue:粤语 cnm:普通话
+        # engine.setProperty('rate', 150)  # 设置语速
+        # engine.setProperty('volume', 1.0)  # 设置音量
+
         for i, (start_end, text) in enumerate(subtitles):
             start_time, end_time = start_end
             duration = end_time - start_time
 
-            tts = gTTS(text=text, lang='zh')
+            # tts = gTTS(text=text, lang='yue')
             audio_path = os.path.join(temp_audio_path, f'subtitle_{i}.mp3')
-            tts.save(audio_path)
+            # tts.save(audio_path)
+            ms_tts_gen(text, audio_path)
             
             audio_segment = AudioSegment.from_mp3(audio_path)
 
@@ -104,6 +112,8 @@ def merge_subtitles(subtitles):
                 break  
 
         current_content = '這裡請注意' + '、'.join(list(set(current_content.replace("細節", "").split('、'))))
+        current_content = current_content.replace("左手、右手", "雙手").replace("右手、左手", "雙手")
+        current_content = current_content.replace("左臂、右臂", "雙臂").replace("右臂、左臂", "雙臂")
         merged_subtitles.append([[current_start, current_end], current_content])
     
     return merged_subtitles