Upload Frondend python file

speak.py

+import os
+import numpy as np
+import librosa
+import tensorflow as tf
+import pyaudio
+import wave
+import threading
+
+from kivy.lang import Builder
+from kivy.uix.screenmanager import Screen, ScreenManager
+from kivymd.app import MDApp
+from kivymd.uix.button import MDRaisedButton
+from kivymd.uix.label import MDLabel
+from main_imports import MDScreen
+from ProjectFiles.applibs import utils
+import warnings
+import pandas as pd
+from sklearn import metrics
+from sklearn.model_selection import train_test_split
+from sklearn.tree import export_graphviz
+from six import StringIO
+import pydotplus
+from sklearn.tree import DecisionTreeClassifier
+
+
+warnings.filterwarnings('ignore')
+utils.load_kv("speak.kv")
+
+
+# Load the saved model
+model = tf.keras.models.load_model('model.h5')
+
+
+
+# Define a function to extract features from audio data
+def extract_features(audio_data, sr=22050):
+    mfcc = librosa.feature.mfcc(y=audio_data, sr=sr, n_mfcc=20)
+    mfcc_mean = np.mean(mfcc, axis=1)
+    mfcc_delta = librosa.feature.delta(mfcc, width=3)
+    mfcc_delta2 = librosa.feature.delta(mfcc, width=3, order=2)
+    mfcc_delta_mean = np.mean(mfcc_delta, axis=1)
+    mfcc_delta2_mean = np.mean(mfcc_delta2, axis=1)
+    features = np.concatenate((mfcc_mean, mfcc_delta_mean, mfcc_delta2_mean))
+    return features
+
+# Define a function to predict the mood from audio data
+def predict_mood(audio_data):
+    features = extract_features(audio_data)
+    features = features.reshape(1, -1)
+    predictions = model.predict(features)
+    emotion_labels = ['angry', 'calm', 'disgust', 'fearful', 'happy', 'neutral', 'sad']
+    predicted_label = emotion_labels[np.argmax(predictions)]
+    return predicted_label
+
+class speak_Screen(Screen):
+
+    def __init__(self, **kwargs):
+        super().__init__(**kwargs)
+        self.recording_thread = None
+        self.recording = False
+        self.audio_buffer = []
+
+    def btnA(self):
+        if not self.recording:
+            self.start_recording()
+            self.ids.l1.text = "Recording started"
+        else:
+            self.ids.l1.text = "Please wait, text is getting ready"
+            self.analyze_audio()
+
+    def start_recording(self):
+        self.recording = True
+        print("Recording started")
+        self.audio_buffer = []  # Clear the audio buffer
+
+        self.recording_thread = threading.Thread(target=self.record_audio)
+        self.recording_thread.start()
+
+    def record_audio(self):
+        chunk = 1024
+        p = pyaudio.PyAudio()
+        stream = p.open(format=pyaudio.paInt16, channels=1, rate=44100, input=True, frames_per_buffer=1024)
+        frames = []
+        while self.recording:
+            data = stream.read(1024)
+            frames.append(data)
+            self.audio_buffer.append(data)
+        stream.stop_stream()
+        stream.close()
+        p.terminate()
+
+    def analyze_audio(self):
+        if not self.audio_buffer:
+            return
+
+        audio_data = np.frombuffer(b''.join(self.audio_buffer), dtype=np.int16)
+
+        # Normalize audio data to the range [-1, 1]
+        audio_data = audio_data.astype(np.float32) / 32768.0
+
+        temp_audio_file = "temp_audio.wav"
+        wf = wave.open(temp_audio_file, 'wb')
+        wf.setnchannels(1)
+        wf.setsampwidth(2)
+        wf.setframerate(44100)
+        wf.writeframes(audio_data.tobytes())
+        wf.close()
+
+        predicted_mood = predict_mood(audio_data)
+        print('Predicted mood:', predicted_mood)
+        self.ids.l1.text = predicted_mood
+