This paper is published in Volume-4, Issue-3, 2018
Area
Speech Processing
Author
Megha Ganeshrao Kadam
Co-authors
Sakshi A. Paithane
Org/Univ
Rajarshi Shahu College of Engineering, Pimpri-Chinchwad, Maharashtra, India
Pub. Date
29 May, 2018
Paper ID
V4I3-1646
Publisher
Keywords
Raspberry Pi 3, Thingspeak cloud server, DC motor driver, Artificial neural network, MFCC, Speech recognition.

Citationsacebook

IEEE
Megha Ganeshrao Kadam, Sakshi A. Paithane. MFCC feature extraction for speech recognition with hybrid application, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
Megha Ganeshrao Kadam, Sakshi A. Paithane (2018). MFCC feature extraction for speech recognition with hybrid application. International Journal of Advance Research, Ideas and Innovations in Technology, 4(3) www.IJARIIT.com.

MLA
Megha Ganeshrao Kadam, Sakshi A. Paithane. "MFCC feature extraction for speech recognition with hybrid application." International Journal of Advance Research, Ideas and Innovations in Technology 4.3 (2018). www.IJARIIT.com.

Abstract

To analyze and detect human voice in different applications such as military area, medical field, and telecommunication for assigning tasks according to it. In this Human voice recognized using MFCC features with a network in such a way that it recognizes only specific person speech commands with exit the program for another one. This paper represents with a wide range of feature extraction algorithm available, MFCC is the is a leading approach for speech feature extraction and our current research aims to apply it on real-time hybrid based applications i.e. home automation and robotic application. The ANN has been trained for commands LIGHTS ON, LIGHTS OFF, FAN ON and FAN OFF for home automation as well as LEFT, RIGHT, FORWARD, BACK, STOP for our robotics application. Thingspeak IOT cloud has been used as the server to send/receive commands between two clients, the PC/laptop from where the speech command is sent and the Raspberry Pi where the command will be used to control the robot and relays for home automation. The best part of the proposed system is that controlling the devices is independent of the location of the speaker. The result shows the proposed method has achieved an accuracy of 96.64% for robotic application and 94.63% for home automation speech commands.
Paper PDF