Update app.py

app.py

@@ -1,5 +1,42 @@
+# import library
 import gradio as gr
-import librosa 
+import librosa
+import pandas as pd
+import numpy as np
 
+import tensorflow as tf
+from tensorflow.keras.layers.experimental import preprocessing
+from tensorflow.keras.preprocessing.image import load_img, img_to_array
+from tensorflow.keras.models import Sequential
+from tensorflow.keras.layers import Dense, Dropout, Flatten, Conv2D, MaxPool2D, BatchNormalization, Input
+
+def get_waveform_label(file):
+    lab = tf.strings.split(file, os.path.sep)[-2]
+    audio_binary = tf.io.read_file(file)
+    audio, _ = tf.audio.decode_wav(audio_binary)
+    waveform=tf.squeeze(audio, axis=-1)
+    return waveform
+
+def get_spectrogram_label(audio):
+    padding = tf.zeros([300000]-tf.shape(audio), dtype=tf.float32)
+    wave = tf.cast(audio, tf.float32)
+    eq_length = tf.concat([wave, padding], 0)
+    spectrogram = tf.signal.stft(eq_length, frame_length=210, frame_step=110)    
+    spectrogram = tf.abs(spectrogram)
+    spectrogram = tf.expand_dims(spectrogram, -1)
+    return spectrogram
+    
+    
 def get_audio(audio):
-  return output
+  audio_waveform = get_waveform_label(audio)
+  audio_spect = get_spectrogram_label(audio_waveform)
+  final_feat = np.array([audio_spect])
+  res = np.argmax(model.predict(final_feat),axis=1)
+  if res == 1:
+    res ="Dog Audio";
+  else:
+    res = "Cat Audio"
+  return res
+  
+  
+ iface = gr.Interface(fn=get_audio,inputs=["audio", "audio"],title="CAT_DOG AUDIO CLASSIFICATION",outputs="html").launch()