Update app.py

app.py

@@ -2,13 +2,9 @@ import gradio as gr
 import numpy as np
 import librosa
 import time
-from transformers import pipeline
 from vaderSentiment.vaderSentiment import SentimentIntensityAnalyzer
 from tensorflow.keras.models import load_model
 
-# Load the ASR pipeline
-p = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-large-960h-lv60-self")
-
 # Load the emotion prediction model
 model = load_model('mymodel_SER_LSTM_RAVDESS.h5')
 
@@ -44,15 +40,14 @@ def sentiment_vader(sentence):
 # Function to transcribe audio and perform sentiment analysis
 def transcribe(audio):
     time.sleep(3)  # Simulate processing delay
-    text = p(audio)["text"]
-    text_sentiment = sentiment_vader(text)
-    return text, text_sentiment
+    # In this case, just return a placeholder value
+    return "Transcription not available"
 
 # Function to get predictions for emotion and sentiment
 def get_predictions(audio_input):
     emotion_prediction = predict_emotion_from_audio(audio_input)
-    transcript, sentiment_prediction = transcribe(audio_input)
-    return emotion_prediction, transcript, sentiment_prediction
+    sentiment_prediction = sentiment_vader(transcribe(audio_input))
+    return emotion_prediction, sentiment_prediction
 
 # Create the Gradio interface
 interface = gr.Interface(
@@ -60,7 +55,6 @@ interface = gr.Interface(
     inputs=gr.Audio(label="Input Audio", type="file"),
     outputs=[
         gr.Label(label="Emotion Prediction"),
-        gr.Textbox(label="Transcript"),
         gr.Label(label="Sentiment Prediction")
     ],
     title="Emotional Machines Test",
@@ -68,4 +62,4 @@ interface = gr.Interface(
 )
 
 # Launch the interface
-interface.launch()
+interface.launch()