Update app.py

app.py

@@ -1,126 +1,65 @@
-import gradio as gr
 import os
-import numpy as np
+import gradio as gr
+import requests
 import joblib
+import numpy as np
 import librosa
-import requests
 from huggingface_hub import hf_hub_download
+from deepface import DeepFace
 
-# --- DeepFace 条件导入 ---
-try:
-    from deepface import DeepFace
-    has_deepface = True
-except ImportError:
-    print("本地未安装 deepface，将在本地跳过臉部情緒；Space 上会安装 deepface。")
-    has_deepface = False
-
-# --- 1. 语音 SVM 加载 ---
-print("Downloading SVM model from Hugging Face Hub...")
-model_path = hf_hub_download(repo_id="GCLing/emotion-svm-model", filename="svm_emotion_model.joblib")
-svm_model = joblib.load(model_path)
-print("SVM model loaded.")
-
-# --- 2. 文本情绪分析：改用 Inference API ---
+# --- 配置：Hugging Face Inference API 文本分析 ---
 HF_API_TOKEN = os.getenv("HF_API_TOKEN")
 if HF_API_TOKEN is None:
-    print("警告：未检测到 HF_API_TOKEN，Inference API 可能失败。")
-# 选用公开存在的中文情感分类模型
+    print("警告：未检测到 HF_API_TOKEN，文字分析可能失败或限流。")
+# 选用公开存在的中文情感分类模型 ID
 HF_TEXT_MODEL = "uer/roberta-base-finetuned-dianping-chinese"
 HF_API_URL = f"https://api-inference.huggingface.co/models/{HF_TEXT_MODEL}"
-headers = {"Authorization": f"Bearer {HF_API_TOKEN}"} if HF_API_TOKEN else {}
-
+HEADERS = {"Authorization": f"Bearer {HF_API_TOKEN}"} if HF_API_TOKEN else {}
 
-def predict_text_via_api(text: str):
-    if not text or text.strip()=="":
+def call_text_api(text: str):
+    if not text or text.strip() == "":
         return {}
     payload = {"inputs": text}
     try:
-        resp = requests.post(HF_API_URL, headers=headers, json=payload, timeout=30)
-        if resp.status_code != 200:
-            print(f"Inference API 返回状态码 {resp.status_code}: {resp.text}")
-            # 退回到简单规则或中性
-            return {"中性": 1.0}
-        data = resp.json()
-        # 根据模型返回格式解析：假设返回 [{"label": "...", "score": ...}, ...]
-        if isinstance(data, list) and len(data)>0 and isinstance(data[0], dict):
-            # 选 top 3 展示
-            result = {}
-            for item in data[:3]:
-                lbl = item.get("label", "")
+        res = requests.post(HF_API_URL, headers=HEADERS, json=payload, timeout=15)
+        res.raise_for_status()
+        data = res.json()
+        result = {}
+        if isinstance(data, list):
+            for item in data:
+                label = item.get("label", "")
                 score = item.get("score", 0.0)
-                # 若标签是英文，可映射到中文；若就是中文可直接用
-                # 例如模型返回 "positive"/"negative"/"neutral"，可映射：
-                if lbl.lower() in ["positive","pos","正面"]:
-                    cn = "正面"
-                elif lbl.lower() in ["negative","neg","负面","負面"]:
-                    cn = "負面"
-                elif lbl.lower() in ["neutral","中性"]:
-                    cn = "中性"
-                else:
-                    cn = lbl
-                result[cn] = float(score)
-            return result
+                result[label] = float(score)
         else:
-            print("Inference API 返回格式异常:", data)
-            return {"中性": 1.0}
+            # 如果返回不同结构，可根据实际调整
+            print("call_text_api 返回格式未预期:", data)
+            return {}
+        return result
     except Exception as e:
-        print("调用 Inference API 出错:", e)
+        print("call_text_api error:", e)
         return {"中性": 1.0}
 
-# 可保留简单规则优先，若规则命中则返回规则，否则调用 API
-emo_keywords = {
-    "happy": ["開心","快樂","愉快","喜悦","喜悅","歡喜","興奮","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"],
-    "disgust": ["噁心","厭惡","反感"]
-}
-negations = ["不","沒","沒有","別","勿","非"]
-def keyword_emotion(text: str):
-    text_proc = text.strip()
-    counts = {emo:0 for emo in emo_keywords}
-    for emo, kws in emo_keywords.items():
-        for w in kws:
-            idx = text_proc.find(w)
-            if idx!=-1:
-                neg=False
-                for neg_word in negations:
-                    plen = len(neg_word)
-                    if idx-plen>=0 and text_proc[idx-plen:idx]==neg_word:
-                        neg=True; break
-                if not neg:
-                    counts[emo]+=1
-    total = sum(counts.values())
-    if total>0:
-        # 归一化并取最高
-        top = max(counts, key=lambda k: counts[k])
-        return {top: counts[top]/total}
-    return None
-
-def predict_text_mixed(text: str):
-    print("predict_text_mixed:", text)
-    if not text or text.strip()=="":
-        return {}
-    res = keyword_emotion(text)
-    if res:
-        # 映射中文标签
-        mapping = {
-            "happy":"高興","angry":"憤怒","sad":"悲傷",
-            "surprise":"驚訝","fear":"恐懼","disgust":"厭惡"
-        }
-        emo = list(res.keys())[0]; prob = float(res[emo])
-        cn = mapping.get(emo, emo)
-        return {cn: prob}
-    # 规则未命中，调用 Inference API
-    return predict_text_via_api(text)
+# --- 语音情绪分析 SVM 模型加载 ---
+USE_VOICE = True
+svm_model = None
+if USE_VOICE:
+    try:
+        print("下载并加载语音 SVM 模型...")
+        model_path = hf_hub_download(repo_id="GCLing/emotion-svm-model", filename="svm_emotion_model.joblib")
+        svm_model = joblib.load(model_path)
+        print("SVM 模型加载完成")
+    except Exception as e:
+        print("语音 SVM 模型加载失败，禁用语音模块:", e)
+        USE_VOICE = False
 
-# --- 3. 语音情绪预测 ---
 def extract_feature(signal: np.ndarray, sr: int) -> np.ndarray:
     mfcc = librosa.feature.mfcc(y=signal, sr=sr, n_mfcc=13)
-    return np.concatenate([np.mean(mfcc, axis=1), np.var(mfcc, axis=1)])
+    feat = np.concatenate([np.mean(mfcc, axis=1), np.var(mfcc, axis=1)])
+    return feat
 
 def predict_voice(audio_path: str):
+    if not USE_VOICE or svm_model is None:
+        return {"error": 1.0}
     if not audio_path:
         return {}
     try:
@@ -133,53 +72,60 @@ def predict_voice(audio_path: str):
         print("predict_voice error:", e)
         return {}
 
-# --- 4. 人脸情绪预测 ---
-import gradio as gr
-
+# --- 臉部情緒分析，使用 DeepFace 分析上傳或拍照圖片 ---
 def predict_face(img: np.ndarray):
-    # 你的 DeepFace 分析逻辑
+    # img 為 numpy array，或 None
     if img is None:
         return {}
-    # ...
-    return {"happy": 0.5, "sad": 0.5}  # 举例
+    try:
+        res = DeepFace.analyze(img, actions=["emotion"], detector_backend="opencv")
+        if isinstance(res, list):
+            first = res[0] if res else {}
+            emo = first.get("emotion", {}) if isinstance(first, dict) else {}
+        else:
+            emo = res.get("emotion", {}) if isinstance(res, dict) else {}
+        emo_fixed = {k: float(v) for k, v in emo.items()}
+        return emo_fixed
+    except Exception as e:
+        print("DeepFace.analyze error:", e)
+        return {}
 
+# --- Gradio 界面 ---
 def build_interface():
     with gr.Blocks() as demo:
-        gr.Markdown("## 多模態情緒分析示例")
+        gr.Markdown("## 多模態情緒分析（簡化版：上傳/拍照人臉 + 語音 + 文字）")
         with gr.Tabs():
-            # 臉部情緒 Tab
+            # 臉部 Tab：上傳或拍照
             with gr.TabItem("臉部情緒"):
-                gr.Markdown("### 臉部情緒 (即時 Webcam Streaming 分析)")
+                gr.Markdown("### 臉部情緒 分析 (上傳或拍照圖片)")
                 with gr.Row():
-                    # 这里用 gr.Image(sources="webcam", streaming=True, type="numpy")
-                    webcam = gr.Image(sources="webcam", streaming=True, type="numpy", label="攝像頭畫面")
-                    face_out = gr.Label(label="情緒分佈")
-                # 每帧送到 predict_face
-                webcam.stream(fn=predict_face, inputs=webcam, outputs=face_out)
+                    # sources=["upload"] 在手機上點上傳可調出相機拍照
+                    face_input = gr.Image(sources=["upload"], type="numpy", label="上傳或拍照圖片")
+                    face_out = gr.Label(label="情緒分布")
+                face_input.change(fn=predict_face, inputs=face_input, outputs=face_out)
 
-            # 語音情緒 Tab
+            # 語音 Tab
             with gr.TabItem("語音情緒"):
-                gr.Markdown("### 語音情緒 分析")
-                with gr.Row():
-                    # 浏览器录音用 source="microphone"
-                    audio = gr.Audio(source="microphone", streaming=False, type="filepath", label="錄音")
-                    voice_out = gr.Label(label="語音情緒結果")
-                audio.change(fn=predict_voice, inputs=audio, outputs=voice_out)
+                gr.Markdown("### 語音情緒 分析 (錄音並上傳)")
+                if USE_VOICE:
+                    with gr.Row():
+                        audio_input = gr.Audio(source="microphone", streaming=False, type="filepath", label="錄音")
+                        voice_out = gr.Label(label="語音情緒結果")
+                    audio_input.change(fn=predict_voice, inputs=audio_input, outputs=voice_out)
+                else:
+                    gr.Markdown("語音模塊不可用。")
 
-            # 文字情緒 Tab
+            # 文字 Tab
             with gr.TabItem("文字情緒"):
-                gr.Markdown("### 文字情緒 分析 (規則+Inference API)")
+                gr.Markdown("### 文字情緒 分析 (Hugging Face Inference API)")
                 with gr.Row():
-                    text = gr.Textbox(lines=3, placeholder="請輸入中文文字…")
+                    text_input = gr.Textbox(lines=3, placeholder="請輸入中文文字…")
                     text_out = gr.Label(label="文字情緒結果")
-                # 使用 submit 触发
-                text.submit(fn=predict_text_mixed, inputs=text, outputs=text_out)
+                text_input.submit(fn=call_text_api, inputs=text_input, outputs=text_out)
 
         return demo
 
-
-
 if __name__ == "__main__":
     demo = build_interface()
-    # share=True 可在本地测试时生成临时公网链接
-    demo.launch(share=True)
+    # share=True 可生成临时公开链接；部署到 Spaces 时无需此参数
+    demo.launch()