|
from __future__ import absolute_import, division, print_function, unicode_literals |
|
|
|
from flask import Flask, make_response, render_template, request, jsonify, redirect, url_for, send_from_directory |
|
from flask_cors import CORS |
|
|
|
import sys |
|
import os |
|
|
|
import librosa |
|
import librosa.display |
|
import numpy as np |
|
|
|
import warnings |
|
import tensorflow as tf |
|
from keras.models import Sequential |
|
from keras.layers import Dense |
|
from keras.utils import to_categorical |
|
from keras.layers import Flatten, Dropout, Activation |
|
from keras.layers import Conv2D, MaxPooling2D |
|
from keras.layers.normalization import BatchNormalization |
|
from sklearn.model_selection import train_test_split |
|
from tqdm import tqdm |
|
|
|
|
|
|
|
import pyaudio |
|
import wave |
|
|
|
warnings.filterwarnings("ignore") |
|
|
|
app = Flask(__name__) |
|
CORS(app) |
|
|
|
classLabels = ('Angry', 'Fear', 'Disgust', 'Happy', 'Sad', 'Surprised', 'Neutral') |
|
numLabels = len(classLabels) |
|
in_shape = (39,216) |
|
model = Sequential() |
|
|
|
model.add(Conv2D(8, (13, 13), input_shape=(in_shape[0], in_shape[1], 1))) |
|
model.add(BatchNormalization(axis=-1)) |
|
model.add(Activation('relu')) |
|
model.add(Conv2D(8, (13, 13))) |
|
model.add(BatchNormalization(axis=-1)) |
|
model.add(Activation('relu')) |
|
model.add(MaxPooling2D(pool_size=(2, 1))) |
|
model.add(Conv2D(8, (3, 3))) |
|
model.add(BatchNormalization(axis=-1)) |
|
model.add(Activation('relu')) |
|
model.add(Conv2D(8, (1, 1))) |
|
model.add(BatchNormalization(axis=-1)) |
|
model.add(Activation('relu')) |
|
model.add(MaxPooling2D(pool_size=(2, 1))) |
|
model.add(Flatten()) |
|
model.add(Dense(64)) |
|
model.add(BatchNormalization()) |
|
model.add(Activation('relu')) |
|
model.add(Dropout(0.2)) |
|
|
|
model.add(Dense(numLabels, activation='softmax')) |
|
model.compile(loss='binary_crossentropy', optimizer='adam', |
|
metrics=['accuracy']) |
|
|
|
|
|
model.load_weights('speech_emotion_detection_ravdess_savee.h5') |
|
|
|
def detect_emotion(file_name): |
|
X, sample_rate = librosa.load(file_name, res_type='kaiser_best',duration=2.5,sr=22050*2,offset=0.5) |
|
sample_rate = np.array(sample_rate) |
|
mfccs = librosa.feature.mfcc(y=X, sr=sample_rate, n_mfcc=39) |
|
feature = mfccs |
|
print("Feature_shape =>",feature.shape) |
|
feature = feature.reshape(39, 216, 1) |
|
result = classLabels[np.argmax(model.predict(np.array([feature])))] |
|
print("Result ==> ",result) |
|
return result |
|
|
|
@app.route("/speech-emotion-recognition/") |
|
def emotion_detection(): |
|
filename = 'audio_files/Happy.wav' |
|
|
|
result = detect_emotion(filename) |
|
return result |
|
|
|
@app.route("/record_audio/") |
|
def record_audio(): |
|
CHUNK = 1024 |
|
FORMAT = pyaudio.paInt16 |
|
CHANNELS = 2 |
|
RATE = 44100 |
|
RECORD_SECONDS = 4 |
|
|
|
fileList = os.listdir('recorded_audio') |
|
print("Audio File List ==> ",fileList) |
|
|
|
new_wav_file = "" |
|
|
|
if(fileList): |
|
filename_list = [] |
|
for i in fileList: |
|
print(i) |
|
filename = i.split('.')[0] |
|
filename_list.append(filename) |
|
|
|
max_file = max(filename_list) |
|
print(type(max_file)) |
|
|
|
new_wav_file = int(max_file) + 1 |
|
else: |
|
new_wav_file="1" |
|
|
|
new_wav_file = str(new_wav_file) + ".wav" |
|
filepath = os.path.join('recorded_audio', new_wav_file) |
|
WAVE_OUTPUT_FILENAME = filepath |
|
|
|
print(WAVE_OUTPUT_FILENAME) |
|
|
|
p = pyaudio.PyAudio() |
|
|
|
stream = p.open(format=FORMAT, |
|
channels=CHANNELS, |
|
rate=RATE, |
|
input=True, |
|
frames_per_buffer=CHUNK) |
|
|
|
print("* recording") |
|
|
|
frames = [] |
|
|
|
for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)): |
|
data = stream.read(CHUNK) |
|
frames.append(data) |
|
|
|
print("* done recording") |
|
|
|
stream.stop_stream() |
|
stream.close() |
|
p.terminate() |
|
|
|
wf = wave.open(WAVE_OUTPUT_FILENAME, 'wb') |
|
wf.setnchannels(CHANNELS) |
|
wf.setsampwidth(p.get_sample_size(FORMAT)) |
|
wf.setframerate(RATE) |
|
wf.writeframes(b''.join(frames)) |
|
wf.close() |
|
return "Audio Recorded" |
|
|
|
if __name__ == "__main__": |
|
app.run() |
|
|
|
|
