✅ Fix: use correct extractor for superb/wav2vec2-base-superb-er

app.py

@@ -21,37 +21,38 @@ def convert_to_wav(uploaded_file):
 
 def get_emotion_label(logits):
     emotions = ["angry", "happy", "neutral", "sad"]
-    scores = torch.softmax(torch.tensor(logits), dim=0).tolist()
+    scores = torch.softmax(logits, dim=0).tolist()
     top_idx = scores.index(max(scores))
     return emotions[top_idx], scores
 
 def analyze_emotion(audio_path):
     extractor, model = load_model()
     waveform, sr = torchaudio.load(audio_path)
+
     if sr != 16000:
-        waveform = torchaudio.transforms.Resample(orig_freq=sr, new_freq=16000)(waveform)
+        waveform = torchaudio.transforms.Resample(sr, 16000)(waveform)
 
-    inputs = extractor(waveform.squeeze().numpy(), sampling_rate=16000, return_tensors="pt")
+    inputs = extractor(waveform[0].numpy(), sampling_rate=16000, return_tensors="pt")
     with torch.no_grad():
         logits = model(**inputs).logits[0]
 
     emotion, scores = get_emotion_label(logits)
     return emotion.capitalize(), scores
 
-# Streamlit UI
+# UI
 st.set_page_config(page_title="🎧 Audio Emotion Detector", layout="centered")
 st.title("🎧 Audio Emotion Analysis (Wav2Vec2)")
 
 uploaded_file = st.file_uploader("Upload an MP3 or WAV audio file", type=["mp3", "wav"])
 
 if uploaded_file:
-    st.audio(uploaded_file, format='audio/wav')
+    st.audio(uploaded_file)
     with st.spinner("Analyzing emotion..."):
         wav_path = convert_to_wav(uploaded_file)
         emotion, scores = analyze_emotion(wav_path)
 
-        st.subheader("Emotion Analysis Result:")
-        st.markdown(f"🧠 **Detected Emotion:** `{emotion}`")
+        st.subheader("Detected Emotion:")
+        st.markdown(f"🧠 **{emotion}**")
 
         st.subheader("Confidence Scores:")
         emotions = ["angry", "happy", "neutral", "sad"]