Create main.py

main.py

@@ -0,0 +1,109 @@
+import tensorflow as tf
+#from transformers import pipeline
+from huggingface_hub import from_pretrained_keras
+import pandas as pd
+import numpy as np
+import joblib
+import os
+import sys
+import pickle
+import shutil
+# 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
+import librosa.display
+import seaborn as sns
+import matplotlib.pyplot as plt
+
+from sklearn.preprocessing import StandardScaler, OneHotEncoder
+from sklearn.metrics import confusion_matrix, classification_report
+from sklearn.model_selection import train_test_split
+
+# to play the audio files
+
+
+import keras
+from keras.preprocessing import sequence
+from keras.models import Sequential, model_from_json
+from keras.layers import Dense, Embedding
+from keras.layers import LSTM, BatchNormalization, GRU
+from keras.preprocessing.text import Tokenizer
+
+from tensorflow.keras.utils import to_categorical
+from keras.layers import Input, Flatten, Dropout, Activation
+from keras.layers import Conv1D, MaxPooling1D, AveragePooling1D
+from keras.models import Model
+from keras.callbacks import ModelCheckpoint
+from tensorflow.keras.optimizers import SGD
+from fastapi import FastAPI, Request, UploadFile, File
+
+
+import warnings
+if not sys.warnoptions:
+    warnings.simplefilter("ignore")
+warnings.filterwarnings("ignore", category=DeprecationWarning)
+
+model=from_pretrained_keras( 'Mohamed41/MODEL_EMOTION_AR_TEXT_72P')
+
+  
+with open('scaler3.pickle', 'rb') as f:
+    scaler3 = pickle.load(f)
+    
+with open('encoder3.pickle', 'rb') as f:
+    encoder3 = pickle.load(f)
+
+def zcr(data,frame_length,hop_length):
+    zcr=librosa.feature.zero_crossing_rate(data,frame_length=frame_length,hop_length=hop_length)
+    return np.squeeze(zcr)
+def rmse(data,frame_length=2048,hop_length=512):
+    rmse=librosa.feature.rms(y=data,frame_length=frame_length,hop_length=hop_length)
+    return np.squeeze(rmse)
+def mfcc(data,sr,frame_length=2048,hop_length=512,flatten:bool=True):
+    mfcc=librosa.feature.mfcc(y=data,sr=sr)
+    return np.squeeze(mfcc.T)if not flatten else np.ravel(mfcc.T)
+
+def extract_features(data,sr=22050,frame_length=2048,hop_length=512):
+    result=np.array([])
+
+    result=np.hstack((result,
+                      zcr(data,frame_length,hop_length),
+                      rmse(data,frame_length,hop_length),
+                      mfcc(data,sr,frame_length,hop_length)
+                     ))
+    return result
+
+def get_predict_feat(path):
+    d, s_rate= librosa.load(path, duration=2.5, offset=0.6)
+    res=extract_features(d)
+    result=np.array(res)
+    result=np.reshape(result,newshape=(1,2376))
+    i_result = scaler3.transform(result)
+    final_result=np.expand_dims(i_result, axis=2)
+
+    return final_result
+
+
+emotions1 = {1: 'Neutral', 2: 'Calm', 3: 'Happy', 4: 'Sad',
+             5: 'Angry', 6: 'Fear', 7: 'Disgust', 8: 'Surprise'}
+
+
+def prediction(path1):
+    res=get_predict_feat(path1)
+    predictions=model.predict(res)
+    y_pred = encoder3.inverse_transform(predictions)
+    print(y_pred[0][0])
+
+
+app = FastAPI()
+
+
+
+
+@app.post("/")
+async def read_root( file: UploadFile = File(...)):
+    file_extension = os.path.splitext(file.filename)[1]
+    with open("tmp"+file_extension, "wb") as buffer:
+        shutil.copyfileobj(file.file, buffer)
+
+     
+    x = prediction("tmp"+file_extension) 
+    return {"filename": file.filename, "filepath": f"/app/{file.filename}","prediction":x}