Update main.py

main.py

@@ -6,6 +6,7 @@ import numpy as np
 import joblib
 import os
 import sys
+import pikle
 
 # librosa is a Python library for analyzing audio and music. It can be used to extract the data from the audio files we will see it later.
 import librosa
@@ -44,32 +45,47 @@ warnings.filterwarnings("ignore", category=DeprecationWarning)
 model=from_pretrained_keras( 'Mohamed41/MODEL_EMOTION_AR_TEXT_72P')
 
 
-def feat_ext(data):
-    # Time_domain_features
+# def feat_ext(data):
+#     # Time_domain_features
+#     # ZCR Persody features or Low level ascoustic features
+#     result = np.array([])
+#     zcr = np.mean(librosa.feature.zero_crossing_rate(y=data).T, axis=0)
+#     result = np.hstack((result, zcr))  # stacking horizontally
+#     # Frequency_domain_features
+#     # Spectral and wavelet Features
+#     # MFCC
+#     mfcc = np.mean(librosa.feature.mfcc(y=data, sr=22050, n_mfcc=40).T, axis=0)
+#     result = np.hstack((result, mfcc))  # stacking horizontally
+#     return result
+
+
+with open('scaler2.pickle', 'rb') as f:
+    scaler = pickle.load(f)
+    
+with open('encoder2.pickle', 'rb') as f:
+    encoder = pickle.load(f)
+
+def feat_ext_test(data):
+    #Time_domain_features
     # ZCR Persody features or Low level ascoustic features
     result = np.array([])
     zcr = np.mean(librosa.feature.zero_crossing_rate(y=data).T, axis=0)
-    result = np.hstack((result, zcr))  # stacking horizontally
-    # Frequency_domain_features
-    # Spectral and wavelet Features
-    # MFCC
-    mfcc = np.mean(librosa.feature.mfcc(y=data, sr=22050, n_mfcc=40).T, axis=0)
-    result = np.hstack((result, mfcc))  # stacking horizontally
+    result=np.hstack((result, zcr)) # stacking horizontally
+    #Frequency_domain_features
+    #Spectral and wavelet Features
+    #MFCC
+    mfcc = np.mean(librosa.feature.mfcc(y=data, sr=22050,n_mfcc=40).T, axis=0)
+    result = np.hstack((result, mfcc)) # stacking horizontally
     return result
 
-
-scaler = joblib.load('scaler.joblib')
-encoder = joblib.load('encoder.joblib')
-
-
 def get_predict_feat(path):
-    d, s_rate = librosa.load(path, duration=2.5, offset=0.6)
-    res = feat_ext(d)
-    result = np.array(res)
-    result = np.reshape(result, newshape=(1, 41))
-    i_result = scaler.transform(result)
-    final_result = np.expand_dims(i_result, axis=2)
-
+    d, s_rate= librosa.load(path, duration=2.5, offset=0.6)
+    res=feat_ext_test(d)
+    result=np.array(res)
+    result=np.reshape(result,newshape=(1,41))
+    i_result = scaler2.transform(result)
+    final_result=np.expand_dims(i_result, axis=2)
+    
     return final_result
 
 
@@ -93,5 +109,5 @@ async def read_root( file: UploadFile = File(...)):
     #     json_data = await request.json()
     # except error:
     #   print (error)   
-      
-    return {"filename": file.filename, "filepath": f"/app/{file.filename}"}
+     x = prediction("/app/{file.filename}") 
+    return {"filename": file.filename, "filepath": f"/app/{file.filename}","prediction()":x}