Signed-off-by: sairate <sairate@sina.cn>

main.py

@@ -180,42 +180,74 @@ def play_mp3(mp3_file):
     while pygame.mixer.music.get_busy():  # 使用一个循环来等待音频播放完毕，保证程序不会在播放结束前退出
         clock.tick(3)
 
+import cv2
+import socket
+import threading
+import queue
+import time
 
-def main():
-    create_connection()
+def detect_face(face_queue):
+    face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_default.xml')
+    video_capture = cv2.VideoCapture(0)  # 打开摄像头 # 用于跟踪是否已经检测到人脸
+    face_detected = False
     while True:
-        # 1. 提示用户发言
-        print('请发言，谢谢！')
-        # 2. 录制音频
-        audio_record(5, 'user_audio.wav')
-
-        print('结束发言')
-        # 3. 获取百度语音识别的access token
-        baidu_token = get_access_token()
-        print('Baidu access token obtained.')
-
-        # 4. 上传录音文件并进行语音识别
-        baidu_result = BaiduYuYin('./user_audio.wav', baidu_token)
-        print('Baidu speech recognition result:', baidu_result)
-
-        # 5. 调用大语言模型进行文本生成
-        model_response = get_completion(baidu_result)
-        print('Model response:', model_response)
-
-        # 6. 将文本转换为语音,保存到唯一的文件名
-        audio_filename="audio/"+str(uuid.uuid4()) + '.mp3'
-        unique_audio_filename = "static/"+audio_filename# 保存为不同的文件名以避免访问冲突
-        asyncio.run(generate_audio_from_text(model_response, unique_audio_filename))
-
-        insert_data(baidu_result, model_response,audio_filename)# 插入数据库
-        # 7. 播放生成的语音
-        play_mp3(unique_audio_filename)
-
-        # 8. 提示用户继续对话或退出
-        user_input = input('继续对话或输入"退出"退出: ')
-        if user_input == '退出':
+        ret, frame = video_capture.read()
+        if not ret:
+            print("无法读取摄像头画面")
             break
 
+        gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
+        faces = face_cascade.detectMultiScale(gray, scaleFactor=1.2, minNeighbors=5)
+
+        # 在图像中绘制检测到的人脸
+        for (x, y, w, h) in faces:
+            cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0), 2)
+
+        # 显示视频流
+        cv2.imshow('Video', frame)
+
+        if len(faces) > 0 and not face_detected:  # 检测到人脸且之前未检测到
+            face_queue.put(True)  # 将检测结果放入队列
+            face_detected = True  # 设置为已检测到人脸
+
+        if len(faces) == 0:  # 如果没有检测到人脸，重置状态
+            face_detected = False
+
+        if cv2.waitKey(1) & 0xFF == ord('q'):  # 按 'q' 键退出
+            break
+
+    video_capture.release()
+    cv2.destroyAllWindows()
+
+def main():
+    face_queue = queue.Queue()
+    detection_thread = threading.Thread(target=detect_face, args=(face_queue,))
+    detection_thread.start()
+
+    while True:
+        if not face_queue.empty():  # 检测到人脸
+            face_result = face_queue.get()
+            if face_result:
+                print("识别到人脸，开始录音...")
+                audio_record(5, 'user_audio.wav')
+                # 获取百度Token
+                baidu_token = get_access_token()
+                baidu_result = BaiduYuYin('./user_audio.wav', baidu_token)
+                print("语音识别结果：",baidu_result)
+                model_response = get_completion(baidu_result)
+                print("模型回复：",model_response)
+                audio_filename = f"audio/{uuid.uuid4()}.mp3"
+                unique_audio_filename = f"static/{audio_filename}"
+                print("开始生成音频文件：", unique_audio_filename)
+                asyncio.run(generate_audio_from_text(model_response, unique_audio_filename))
+                print("数据库信息",baidu_result, model_response, unique_audio_filename)
+                insert_data(baidu_result, model_response, audio_filename)
+                play_mp3(unique_audio_filename)
+                face_queue.put(False)
+                if input("请输入（退出）")=="退出":
+                    break
+
+    detection_thread.join()  # 等待人脸检测线程结束
 
 if __name__ == "__main__":
     main()

templates/index.html

@@ -10,7 +10,7 @@
             font-family: Arial, sans-serif;
             margin: 0;
             padding: 20px;
-            background-color: #f0f0f0;
+            background-color: #cbe5d6;
         }
         h1 {
             text-align: center;
@@ -24,11 +24,11 @@
         th, td {
             padding: 12px;
             text-align: left;
-            border-bottom: 1px solid #ddd;
+            border-bottom: 1px solid #9714b9;
         }
         th {
             background-color: #5e7abf;
-            color: white;
+            color: #278080;
         }
         tr:nth-child(even) {
             background-color: #8568d7;

test.py

@@ -7,5 +7,33 @@ def get_completion(prompt, model="solar"):
             'content': prompt,
         },
     ])
+    print(response)
+    print(response['message'])
     return response['message']['content']
-print(get_completion("你好，我是小明。"))
+print(get_completion("你好，我是小明。"))
+# import cv2
+
+# def test_camera():
+#     video_capture = cv2.VideoCapture(0)  # 打开默认摄像头
+#
+#     if not video_capture.isOpened():
+#         print("无法打开摄像头")
+#         return
+#
+#     while True:
+#         ret, frame = video_capture.read()
+#         if not ret:
+#             print("无法读取摄像头画面")
+#             break
+#
+#         cv2.imshow('Video', frame)
+#
+#         if cv2.waitKey(1) & 0xFF == ord('q'):  # 按 'q' 键退出
+#             break
+#
+#     video_capture.release()
+#     cv2.destroyAllWindows()
+#
+# # 调用函数
+# test_camera()
+