Update app.py

app.py

@@ -210,38 +210,48 @@ def analyze_voice(audio_data, state):
         return state, "음성을 먼저 녹음해주세요.", "", "", ""
 
     try:
-        if isinstance(audio_data, tuple):
-            sr, y = audio_data
-        elif isinstance(audio_data, str):
-            y, sr = librosa.load(audio_data, sr=16000)
-        else:
-            print("Unsupported audio format")
-            return state, "오디오 형식을 지원하지 않습니다.", "", "", ""
+        # 오디오 데이터 처리
+        sr, y = audio_data  # 직접 튜플 언패킹
+        
+        if len(y) == 0:
+            return state, "음성이 감지되지 않았습니다.", "", "", ""
 
         # 음향학적 특성 분석
-        acoustic_features = calculate_baseline_features((sr, y))
-        if acoustic_features is None:
-            return state, "음성 분석에 실패했습니다.", "", "", ""
-
-        # 음성 감정 분석
-        voice_emotion = map_acoustic_to_emotion(acoustic_features, state.get("baseline_features"))
+        acoustic_features = {
+            "energy": float(np.mean(librosa.feature.rms(y=y))),
+            "tempo": float(librosa.beat.tempo(y=y, sr=sr)[0]),
+            "pitch": float(np.mean(librosa.feature.zero_crossing_rate(y=y))),
+            "volume": float(np.mean(np.abs(y)))
+        }
 
         # 음성 인식
         if speech_recognizer:
-            transcription = speech_recognizer({"sampling_rate": sr, "raw": y})
-            text = transcription["text"]
+            try:
+                transcription = speech_recognizer({"sampling_rate": sr, "raw": y.astype(np.float32)})
+                text = transcription["text"]
+            except Exception as e:
+                print(f"Speech recognition error: {e}")
+                text = "음성 인식 실패"
         else:
             text = "음성 인식 모델을 불러올 수 없습니다."
 
+        # 음성 감정 분석
+        voice_emotion = map_acoustic_to_emotion(acoustic_features, state.get("baseline_features"))
+
         # 텍스트 감정 분석
         if text_analyzer and text:
-            text_sentiment = text_analyzer(text)[0]
-            text_result = f"텍스트 감정 분석: {text_sentiment['label']} (점수: {text_sentiment['score']:.2f})"
+            try:
+                text_sentiment = text_analyzer(text)[0]
+                text_result = f"텍스트 감정 분석: {text_sentiment['label']} (점수: {text_sentiment['score']:.2f})"
+            except Exception as e:
+                print(f"Text analysis error: {e}")
+                text_sentiment = {"label": "unknown", "score": 0.0}
+                text_result = "텍스트 감정 분석 실패"
         else:
             text_sentiment = {"label": "unknown", "score": 0.0}
             text_result = "텍스트 감정 분석을 수행할 수 없습니다."
 
-        # 결과 포맷팅
+        # 결과 포맷팅 
         voice_result = (
             f"음성 감정: {voice_emotion['primary']} "
             f"(강도: {voice_emotion['intensity']:.1f}%, 신뢰도: {voice_emotion['confidence']:.2f})\n"
@@ -255,11 +265,10 @@ def analyze_voice(audio_data, state):
 
         # 프롬프트 생성
         prompt = generate_detailed_prompt(text, voice_emotion, text_sentiment)
-
-        # 상태 업데이트
         state = {**state, "final_prompt": prompt}
 
         return state, text, voice_result, text_result, prompt
+
     except Exception as e:
         print(f"Error in analyze_voice: {str(e)}")
         return state, f"오류 발생: {str(e)}", "", "", ""
@@ -322,16 +331,24 @@ def generate_image_from_prompt(prompt):
         print(f"Error generating image: {str(e)}")
         return None
 
-def save_reflection(text, state):
-    """감상 저장"""
+def save_reflection_fixed(text, state):
     if not text.strip():
         return state, []
-
+    
     try:
         current_time = datetime.now().strftime("%H:%M:%S")
         if text_analyzer:
             sentiment = text_analyzer(text)[0]
             sentiment_text = f"{sentiment['label']} ({sentiment['score']:.2f})"
+            
+            # 감정 분석 결과를 프롬프트에 반영
+            emotion_prompt = generate_detailed_prompt(
+                text, 
+                {"primary": sentiment['label'], "intensity": sentiment['score'] * 100, 
+                 "characteristics": ["텍스트 기반 감정"], "confidence": sentiment['score']},
+                sentiment
+            )
+            state = {**state, "final_prompt": emotion_prompt}
         else:
             sentiment_text = "분석 불가"
 
@@ -341,8 +358,8 @@ def save_reflection(text, state):
         state = {**state, "reflections": reflections}
         return state, reflections
     except Exception as e:
-        print(f"Error in save_reflection: {str(e)}")
-        return state, []
+        print(f"Error saving reflection: {e}")
+        return state, state.get("reflections", [])
 
 def create_interface():
     db = SimpleDB()
@@ -383,7 +400,8 @@ def create_interface():
                 baseline_audio = gr.Audio(
                     label="축원 문장 녹음하기",
                     sources=["microphone"],
-                    type="numpy"
+                    type="numpy",
+                    streaming=False
                 )
                 set_baseline_btn = gr.Button("기준점 설정 완료", variant="primary")
                 baseline_status = gr.Markdown("")
@@ -422,7 +440,8 @@ def create_interface():
                         voice_input = gr.Audio(
                             label="소원을 나누고 싶은 마음을 말해주세요",
                             sources=["microphone"],
-                            type="numpy"
+                            type="numpy",
+                            streaming=False
                         )
                         with gr.Row():
                             clear_btn = gr.Button("녹음 지우기", variant="secondary")
@@ -480,15 +499,16 @@ def create_interface():
                 return (
                     "이름을 입력해주세요", 
                     gr.update(visible=False), 
-                    current_state
+                    current_state,
+                    gr.update(selected=0)  # 현재 탭 유지
                 )
             current_state = {**current_state, "user_name": name}
             return (
                 WORLDVIEW_MESSAGE,
                 gr.update(visible=True),
-                current_state
+                current_state,
+                gr.update(selected=1)  # 축원 탭(두 번째 탭)으로 이동
             )
-
         def handle_baseline(audio, current_state):
             if audio is None:
                 return current_state, "음성을 먼저 녹음해주세요."
@@ -570,7 +590,7 @@ def create_interface():
         start_btn.click(
             fn=handle_start,
             inputs=[name_input, state],
-            outputs=[worldview_display, worldview_display, state]
+            outputs=[worldview_display, worldview_display, state, tabs]
         )
 
         set_baseline_btn.click(