Last commit not found
| 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 | |
| from datetime import date | |
| import re | |
| import json | |
| import email | |
| import csv | |
| import datetime | |
| import smtplib | |
| import ssl | |
| from email.mime.text import MIMEText | |
| import time | |
| import pytz | |
| import requests | |
| # import pyaudio | |
| import wave | |
| import shutil | |
| import warnings | |
| import tensorflow as tf | |
| import gradio as gr | |
| 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 import BatchNormalization | |
| from sklearn.model_selection import train_test_split | |
| from tqdm import tqdm | |
| warnings.filterwarnings("ignore") | |
| timestamp = datetime.datetime.now() | |
| current_date = timestamp.strftime('%d-%m-%Y') | |
| current_time = timestamp.strftime('%I:%M:%S') | |
| IP = '' | |
| cwd = os.getcwd() | |
| 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']) | |
| # print(model.summary(), file=sys.stderr) | |
| model.load_weights('speech_emotion_detection_ravdess_savee.h5') | |
| # app = Flask(__name__) | |
| # app._static_folder = os.path.join( "/home/ubuntu/Desktop/nlpdemos/server_demos/speech_emotion/static" ) | |
| def selected_audio(audio): | |
| if audio and audio != 'Please select any of the following options': | |
| post_file_name = audio.lower() + '.wav' | |
| filepath = os.path.join("pre_recoreded",post_file_name) | |
| if os.path.exists(filepath): | |
| print("SELECT file name => ",filepath) | |
| result = predict_speech_emotion(filepath) | |
| print("result = ",result) | |
| return result | |
| def recorded_audio(audio): | |
| try: | |
| fileList = os.listdir('recorded_audio') | |
| new_wav_file = "" | |
| if(fileList): | |
| filename_list = [] | |
| for i in fileList: | |
| filename = i.split('.')[0] | |
| filename_list.append(int(filename)) | |
| max_file = max(filename_list) | |
| 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) | |
| # shutil.move(recorded_audio, filepath) | |
| filepath = 'recorded_audio/22.wav' | |
| result = predict_speech_emotion(audio.name) | |
| return result | |
| except Exception as e: | |
| print(e) | |
| return "ERROR" | |
| def predict_speech_emotion(filepath): | |
| if os.path.exists(filepath): | |
| print("last file name => ",filepath) | |
| X, sample_rate = librosa.load(filepath, 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 | |
| feature = feature.reshape(39, 216, 1) | |
| # np_array = np.array([feature]) | |
| np_array = np.array([feature]) | |
| prediction = model.predict(np_array) | |
| np_argmax = np.argmax(prediction) | |
| result = classLabels[np_argmax] | |
| return result | |
| # demo = gr.Interface( | |
| # fn=send_audio, | |
| # inputs=gr.Audio(source="microphone", type="filepath"), | |
| # outputs="text") | |
| # demo.launch() | |
| # selected_audio = gr.Dropdown(["Angry", "Happy", "Sad", "Disgust","Fear", "Surprise", "Neutral"], | |
| # lable = "Input Audio") | |
| # audio_ui=gr.Audio() | |
| # text = gr.Textbox() | |
| # demo = gr.Interface( | |
| # fn=send_audio, | |
| # inputs=selected_audio, | |
| # outputs=[audio_ui,text]) | |
| # demo.launch() | |
| def return_audio_clip(audio_text): | |
| post_file_name = audio_text.lower() + '.wav' | |
| filepath = os.path.join("pre_recoreded",post_file_name) | |
| return filepath | |
| with gr.Blocks(css=".gradio-container {background-color: lightgray;}") as demo: | |
| gr.Markdown("Select audio or record audio") | |
| with gr.Row(): | |
| with gr.Column(): | |
| input_audio_text = gr.Dropdown(lable="Input Audio",choices=["Please select any of the following options","Angry", "Happy", "Sad", "Disgust","Fear", "Surprise", "Neutral"],interactive=True) | |
| audio_ui=gr.Audio() | |
| input_audio_text.change(return_audio_clip,input_audio_text,audio_ui) | |
| output_text = gr.Textbox(lable="Prdicted emotion") | |
| sub_btn = gr.Button("Submit") | |
| with gr.Column(): | |
| audio=gr.Audio(labele="Recored audio",source="microphone", type="file") | |
| recorded_text = gr.Textbox(lable="Prdicted emotion") | |
| with gr.Column(): | |
| sub_btn2 = gr.Button("Submit") | |
| gr.Markdown("""Feel free to give us your [feedback](https://www.pragnakalp.com/contact/) and contact us at [[email protected]]("mailto:[email protected]") | |
| if you want to have your own Speech emotion detection system. We are just one click away. And don't forget to check out more interesting | |
| [NLP services](https://www.pragnakalp.com/services/natural-language-processing-services/) we are offering. | |
| <p style='text-align: center;'>Developed by :<a href="https://www.pragnakalp.com" target="_blank"> Pragnakalp Techlabs</a></p>""") | |
| sub_btn.click(selected_audio, inputs=input_audio_text, outputs=output_text) | |
| sub_btn2.click(recorded_audio, inputs=audio, outputs=recorded_text) | |
| demo.launch() |