Create app.py

app.py

@@ -0,0 +1,49 @@
+p = pipeline("automatic-speech-recognition")
+
+from tensorflow.keras.models import load_model
+
+model = load_model('/content/drive/MyDrive/Mis ML Modelos/mymodel_SER_LSTM_RAVDESS.h5')
+
+def extract_mfcc(wav_file_name):
+    #This function extracts mfcc features and obtain the mean of each dimension
+    y, sr = librosa.load(wav_file_name)
+    mfccs = np.mean(librosa.feature.mfcc(y=y, sr=sr, n_mfcc=40).T,axis=0)
+    return mfccs
+
+emotions = {1: 'neutral', 2: 'calm', 3: 'happy', 4: 'sad', 5: 'angry', 6: 'fearful', 7: 'disgust', 8: 'surprised'}
+
+def predict_emotion_from_audio(wav_filepath):
+    test_point = extract_mfcc(wav_filepath)
+    test_point = np.reshape(test_point, newshape=(1, 40, 1))
+    predictions = model.predict(test_point)
+    predicted_emotion_label = np.argmax(predictions[0]) + 1
+    return emotions[predicted_emotion_label]
+
+# Semantics
+def sentiment_vader(sentence):
+
+    # Create a SentimentIntensityAnalyzer object.
+    sid_obj = SentimentIntensityAnalyzer()
+
+    sentiment_dict = sid_obj.polarity_scores(sentence)
+    negative = sentiment_dict['neg']
+    neutral = sentiment_dict['neu']
+    positive = sentiment_dict['pos']
+    compound = sentiment_dict['compound']
+
+    if sentiment_dict['compound'] >= 0.05 :
+        overall_sentiment = "Positive"
+
+    elif sentiment_dict['compound'] <= - 0.05 :
+        overall_sentiment = "Negative"
+
+    else :
+        overall_sentiment = "Neutral"
+
+    return overall_sentiment
+
+def transcribe(audio, state=""):
+    time.sleep(3)
+    text = p(audio)["text"]
+    text = sentiment_vader(text)
+    return text