Fix/smooth mouth (#412)

* 1. 修复了musetalk方案中，当数字人说话状态变化时，嘴部画面跳变问题；
2. 新增现代美观的前端dashboard.html，集成了对话与朗读功能；
3. 修复了“'weights_only' is an invalid keyword argument for load()”报错。

* bugfix:修复视频连接状态不更新的bug

* feature:新增可选是否启用musereal中的混合过度选项

* 参照fix log修复log

---------

Co-authored-by: marstaos <liu.marstaos@outlook.com>

.gitignore

@@ -16,6 +16,10 @@ pretrained
 .DS_Store
 workspace/log_ngp.txt
 .idea
+keep_gpu.py
+models/
+*.log
+prepared_data/
 models/
 *.log
 dist

musereal.py

@@ -267,7 +267,9 @@ class MuseReal(BaseReal):
       
 
     def process_frames(self,quit_event,loop=None,audio_track=None,video_track=None):
-        # 新增状态跟踪变量
+        enable_transition = True  # 设置为False禁用过渡效果，True启用
+        
+        if enable_transition:
             self.last_speaking = False
             self.transition_start = time.time()
             self.transition_duration = 0.1  # 过渡时间
@@ -280,6 +282,7 @@ class MuseReal(BaseReal):
             except queue.Empty:
                 continue
             
+            if enable_transition:
                 # 检测状态变化
                 current_speaking = not (audio_frames[0][1]!=0 and audio_frames[1][1]!=0)
                 if current_speaking != self.last_speaking:
@@ -297,6 +300,7 @@ class MuseReal(BaseReal):
                 else:
                     target_frame = self.frame_list_cycle[idx]
                 
+                if enable_transition:
                     # 说话→静音过渡
                     if time.time() - self.transition_start < self.transition_duration and self.last_speaking_frame is not None:
                         alpha = min(1.0, (time.time() - self.transition_start) / self.transition_duration)
@@ -305,6 +309,8 @@ class MuseReal(BaseReal):
                         combine_frame = target_frame
                     # 缓存静音帧
                     self.last_silent_frame = combine_frame.copy()
+                else:
+                    combine_frame = target_frame
             else:
                 self.speaking = True
                 bbox = self.coord_list_cycle[idx]
@@ -318,8 +324,9 @@ class MuseReal(BaseReal):
                 mask = self.mask_list_cycle[idx]
                 mask_crop_box = self.mask_coords_list_cycle[idx]
 
-                # 静音→说话过渡
                 current_frame = get_image_blending(ori_frame,res_frame,bbox,mask,mask_crop_box)
+                if enable_transition:
+                    # 静音→说话过渡
                     if time.time() - self.transition_start < self.transition_duration and self.last_silent_frame is not None:
                         alpha = min(1.0, (time.time() - self.transition_start) / self.transition_duration)
                         combine_frame = cv2.addWeighted(self.last_silent_frame, 1-alpha, current_frame, alpha, 0)
@@ -327,6 +334,8 @@ class MuseReal(BaseReal):
                         combine_frame = current_frame
                     # 缓存说话帧
                     self.last_speaking_frame = combine_frame.copy()
+                else:
+                    combine_frame = current_frame
 
             image = combine_frame
             new_frame = VideoFrame.from_ndarray(image, format="bgr24")