To search, Click below search items. |
| All Published Papers Search Service |
|
|
|
Title |
Real-Time Implementation of Isolated-Word Speech Recognition System on Raspberry Pi 3 Using WAT-MFCC |
|
Author |
Mohamed Walid, Souha Bousselmi, Karim Dabbabi and Adnen Cherif |
| Citation |
Vol. 19 No. 3 pp. 42-49 |
|
Abstract |
The automatic speech recognition (ASR) is a field of research that has been emerged in the early 1950s, and has been used in the literature for convenient and efficient human-machine interaction. For voice command system, it is based on implementation of isolated-word speech recognition and it can include many applications, such as voice-activated devices, robots, access control system, etc. In this paper, such isolated-word speech recognition system has been implemented on Raspberry-Pi 3 (RPi 3) board by combining Wave Atoms Transform (WAT) approach and Frequency-Mel Cepstral Coefficients (MFCC) (WAT-MFCC) with Support Vector Machine (SVM). The experiments have been tested on Arabic words database and the achieved results have proven the reliability of the proposed WAT-MFCC-SVM recognition approach with a rate of 100% and a Real-Time Factor (RTF) of 1.50. |
|
Keywords |
Isolated-word speech recognition, Real-Time implementation, WAT, MFCC, WAT-MFCC, SVM, Raspberry PI 3 board. |
|
URL |
