Update app.py

app.py

@@ -41,16 +41,18 @@ def extract_feature(signal: np.ndarray, sr: int) -> np.ndarray:
 # --- 4. 三種預測函式 ---
 
 def predict_face(img: np.ndarray):
-    if img is None:
-        return {}  # 没有帧时返回空
+    print("predict_face called, img is None?", img is None)
+    # 你的限频和 DeepFace 分析逻辑...
     try:
         result = DeepFace.analyze(img, actions=["emotion"], detector_backend="opencv")
-        return result.get("emotion", {})
+        emo = result.get("emotion", {})
+        print("DeepFace result:", emo)
+        return emo
     except Exception as e:
-        # 遇到错误时，可返回空或日志
-        print("DeepFace 分析错误：", e)
+        print("DeepFace.analyze error:", e)
         return {}
 
+
 def predict_voice(audio):
     """
     語音情緒分析：audio 由 Gradio 傳入，形式為暫存檔路徑字串 (str)。
@@ -65,15 +67,20 @@ def predict_voice(audio):
     return {labels[i]: float(probs[i]) for i in range(len(labels))}
 
 def predict_text(text: str):
-    """
-    文字情緒分析：使用 transformers pipeline，
-    輸入中文字串，回傳 dict，例如 {"POSITIVE":0.95} 或模型輸出標籤與信心分數。
-    """
-    if not text or text.strip() == "":
+    
+def predict_text(text: str):
+    print("predict_text called, text:", text)
+    if not text or text.strip()=="":
         return {}
-    pred = text_emotion(text)[0]
-    # pred 形如 {"label": "...", "score": ...}
-    return {pred["label"]: float(pred["score"])}
+    try:
+        pred = text_emotion(text)[0]
+        result = {pred["label"]: float(pred["score"])}
+        print("Text sentiment result:", result)
+        return result
+    except Exception as e:
+        print("predict_text error:", e)
+        return {}
+
 
 # --- 5. 建立 Gradio 介面 ---
 with gr.Blocks() as demo:
@@ -85,9 +92,9 @@ with gr.Blocks() as demo:
             with gr.Row():
                 webcam = gr.Image(sources="webcam", streaming=True, type="numpy", label="攝像頭畫面")
                 emotion_output = gr.Label(label="情緒分布")
-            # 关键：用 stream 让每帧到达时调用 predict_face 并更新 emotion_output
             webcam.stream(fn=predict_face, inputs=webcam, outputs=emotion_output)
 
+
         # 其餘 Tab 可按原先寫法，或用 Blocks 方式
         with gr.TabItem("語音情緒"):
             audio = gr.Audio(sources="microphone", streaming=False, type="filepath", label="錄音")
@@ -98,7 +105,9 @@ with gr.Blocks() as demo:
         with gr.TabItem("文字情緒"):
             text = gr.Textbox(lines=3, placeholder="請輸入中文文字…")
             text_output = gr.Label(label="文字情緒結果")
-            text.submit(fn=predict_text, inputs=text, outputs=text_output)
+            btn = gr.Button("分析文字")
+            btn.click(fn=predict_text, inputs=text, outputs=text_output)
+