app.py

import os
import librosa
import numpy as np
import tensorflow as tf
import gradio as gr
from sklearn.preprocessing import LabelEncoder
import pandas as pd

# Load the pre-trained model
model_path = 'sound_to_text_model.h5'
model = tf.keras.models.load_model(model_path)

# Load the labels from labels.csv
labels_df = pd.read_csv('labels.csv')  # Make sure this path is correct
training_labels = labels_df['Label'].tolist()  # Assuming the column name is 'Label'

# Initialize the encoder and fit it to your labels
encoder = LabelEncoder()
encoder.fit(training_labels)  # Fit the encoder to your labels

# Function to extract features from audio
def extract_features(file_path):
    y_audio, sr = librosa.load(file_path, duration=2.0)
    mfccs = librosa.feature.mfcc(y=y_audio, sr=sr, n_mfcc=13)
    return np.mean(mfccs.T, axis=0)  # Average to create a fixed size

# Function to predict text from audio
def predict_sound_text(audio):
    features = extract_features(audio)  # Use audio directly as the file path
    prediction = model.predict(np.array([features]))
    label = encoder.inverse_transform([np.argmax(prediction)])
    return label[0]

# Define Gradio interface
interface = gr.Interface(
    fn=predict_sound_text,
    inputs=gr.Audio(type="filepath"),  # Use only the type argument
    outputs="text",
    title="Audio to Text Converter",
    description="Upload an audio file (MP3 format) and get the textual representation."
)

# Launch the interface
if __name__ == "__main__":
    interface.launch()