修改is_speaking状态

app.py

@@ -217,7 +217,7 @@ async def human(request):
             nerfreals[sessionid].flush_talk()
 
         if params['type'] == 'echo':
-            nerfreals[sessionid].speaking = True
+            nerfreals[sessionid].liv_speaking = True
             nerfreals[sessionid].put_msg_txt(params['text'])
 
         elif params['type'] == 'chat':

basereal.py

@@ -91,6 +91,8 @@ class BaseReal:
 
         self.speaking = False
 
+        self.liv_speaking = False
+
         self.recording = False
         self._record_video_pipe = None
         self._record_audio_pipe = None
@@ -141,7 +143,7 @@ class BaseReal:
         self.asr.flush_talk()
 
     def is_speaking(self)->bool:
-        return self.speaking
+        return self.liv_speaking
 
     def __loadcustom(self):
         for item in self.opt.customopt:
@@ -163,6 +165,13 @@ class BaseReal:
 
     def notify(self, eventpoint):
         logger.info("notify:%s", eventpoint)
+        # 使用字典的键访问方式，而不是对象属性方式
+        if eventpoint['status'] == "start":
+            self.liv_speaking = True
+            logger.info("tts start")  # 这里可能是笔误，改为start更合理
+        if eventpoint['status'] == "end":
+            self.liv_speaking = False
+            logger.info("tts end")
 
     def start_recording(self):
         """开始录制视频"""

ttsreal.py

@@ -253,7 +253,8 @@ class SovitsTTS(BaseTTS):
             'prompt_text':reftext,
             'prompt_lang':language,
             'media_type':'ogg',
-            'streaming_mode':True
+            'streaming_mode':True,
+            "speed_factor":1.2
         }
         # req["text"] = text
         # req["text_language"] = language