Update app.py

app.py

@@ -1,61 +1,77 @@
+# app.py
+import os
+# ─── 解決 DeepFace 無法寫入預設路徑的問題 ─────────────────────────────────────────
+# 將 DeepFace 的快取目錄指向可寫入的 /tmp 之下
+os.environ["DEEPFACE_HOME"] = "/tmp/.deepface"
+
 import gradio as gr
 import numpy as np
-import joblib, io, base64
+import joblib, io
 import librosa
 from deepface import DeepFace
 
-# —— 1. 載入模型 —— 
-#    如果只要上傳 wav/text，就先把 face 部分包在 fn 裡動態 load
-audio_model = joblib.load("src/voice_model.joblib")
+# ─── 1. 載入模型 ─────────────────────────────────────────────────────────────
+# 我們把模型檔放在跟 app.py 同一層的 voice_model.joblib
+MODEL_PATH = os.path.join(os.path.dirname(__file__), "voice_model.joblib")
+audio_model = joblib.load(MODEL_PATH)
 
+# ─── 2. 定義各種分析函數 ────────────────────────────────────────────────────
 def analyze_face(frame: np.ndarray):
     # DeepFace 回傳 dict，裡面有 'dominant_emotion'
     res = DeepFace.analyze(frame, actions=["emotion"], enforce_detection=False)
-    emo = res["dominant_emotion"]
-    return frame, emo
+    return frame, res["dominant_emotion"]
 
 def analyze_audio(wav_file):
     wav_bytes = wav_file.read()
+    # 用 librosa 讀入
     y, sr = librosa.load(io.BytesIO(wav_bytes), sr=None)
     mfccs = librosa.feature.mfcc(y=y, sr=sr, n_mfcc=13)
     mf = np.mean(mfccs.T, axis=0)
     return audio_model.predict([mf])[0]
 
 def analyze_text(txt):
+    # 簡單關鍵字 mapping
     mapping = {
-        "😊happy":["開心","快樂","愉快","喜悅","歡喜","興奮","歡","高興"],
-        "😠angry": ["生氣","憤怒","不爽","發火","火大","氣憤"],
-        "😢sad":   ["傷心","難過","哭","難受","心酸","憂","悲","哀","痛苦","慘","愁"],
-        "😲surprise": ["驚訝","意外","嚇","驚詫","詫異","訝異","好奇"],
-        "😨fear":  ["怕","恐懼","緊張","懼","膽怯","畏"],
+        "😊 happy":   ["開心","快樂","愉快","喜悅","歡喜","興奮","高興","歡"],
+        "😠 angry":   ["生氣","憤怒","不爽","發火","火大","氣憤"],
+        "😢 sad":     ["傷心","難過","哭","憂","悲","心酸","哀","痛苦","慘","愁"],
+        "😲 surprise":["驚訝","意外","嚇","好奇","驚詫","詫異","訝異"],
+        "😨 fear":    ["怕","恐懼","緊張","懼","膽怯","畏"],
     }
     for emo, kws in mapping.items():
         if any(w in txt for w in kws):
             return emo
-    return "neutral"
+    return "😐 neutral"
 
-with gr.Blocks() as demo:
-    gr.Markdown("## 多模態即時情緒分析")
+# ─── 3. 建立 Gradio 介面 ────────────────────────────────────────────────────
+with gr.Blocks(title="多模態即時情緒分析") as demo:
+    gr.Markdown("## 🤖 多模態即時情緒分析")
     with gr.Tabs():
         with gr.TabItem("📷 Live Face"):
-            from gradio import components as grc
-
-            camera = grc.Camera(label="請對準鏡頭")
-
-            out_img = gr.Image(label="畫面")
-            out_label = gr.Label(label="情緒")
-            camera.change(fn=analyze_face, inputs=camera, outputs=[out_img, out_label], live=True)
+            # 注意要用 gr.components.Camera 或 gr.components…
+            camera = gr.components.Camera(label="請對準鏡頭 (Live)")
+            out_img = gr.Image(label="擷取畫面")
+            out_lbl = gr.Label(label="檢測到的情緒")
+            camera.change(
+                fn=analyze_face,
+                inputs=camera,
+                outputs=[out_img, out_lbl],
+                live=True
+            )
 
-        with gr.TabItem("🎤 上傳語音"):
-            wav = gr.File(label="選擇 .wav 檔", file_types=[".wav"])
-            wav_btn = gr.Button("分析")
-            wav_out = gr.Text(label="偵測到的情緒")
+        with gr.TabItem("🎤 上傳語音檔"):
+            wav = gr.File(label="選擇 .wav 檔案", file_types=[".wav"])
+            wav_btn = gr.Button("開始分析")
+            wav_out = gr.Textbox(label="語音偵測到的情緒")
             wav_btn.click(fn=analyze_audio, inputs=wav, outputs=wav_out)
 
         with gr.TabItem("⌨️ 輸入文字"):
             txt = gr.Textbox(label="在此輸入文字", lines=3)
-            txt_btn = gr.Button("分析")
-            txt_out = gr.Text(label="偵測到的情緒")
+            txt_btn = gr.Button("開始分析")
+            txt_out = gr.Textbox(label="文字偵測到的情緒")
             txt_btn.click(fn=analyze_text, inputs=txt, outputs=txt_out)
 
+# 啟動
+if __name__ == "__main__":
+    # Hugging Face Spaces 上不需要傳 host/port，直接 .launch() 即可
     demo.launch()