Spoken language identification based on optimised genetic algorithm–extreme learning machine approach

Musatafa Abbas Abbood Albadr, Sabrina Tiun, Masri Ayob, Fahad Taha AL-Dhief

Research output: Contribution to journalArticle

Abstract

The determination and classification of a recognized spoken language based on certain contents and datasets is known as the process of language identification (LID). The common process in carrying out LID entails the mandatory processing of data which enables the extraction of the necessary features for the process. The extraction involves a mature process whereby the development of the standard LID features have been conducted much earlier by means of a mel-frequency cepstral coefficients, shifted delta cepstral, Gaussian mixture model and i-vector-based framework. Despite that, improvement or rather optimisation still needs to be done on the learning process based on the extracted features so as to obtain all the knowledge embedded within them. The classification and regression analysis can benefit tremendously from the use of the extreme learning machine (ELM) which is a particularly effective and useful learning model for training a single-hidden layer neural network. However, owing to the randomly selected weights embedded in the input’s hidden layers, the model’s learning process is rendered to be ineffective or not optimised in its entirety. In this study, the ELM is employed as the learning model for LID due to the standard feature extraction. In addition, this study proposes a new optimised genetic algorithm (OGA) with three different selection criteria (i.e., roulette wheel, K-tournament and random) to select the appropriate initial weights and biases of the input hidden layer of the ELM, thereby minimising the classification error and improving the general performance of the ELM for LID. Results show the excellent performance of the proposed OGA–ELM with three different selection criteria, namely, roulette wheel, K-tournament and random, with the highest accuracies of 99.50%, 100% and 99.38%, respectively.

Original languageEnglish
JournalInternational Journal of Speech Technology
DOIs
Publication statusPublished - 1 Jan 2019

Fingerprint

spoken language
Learning systems
Identification (control systems)
learning
Wheels
language
learning process
Regression analysis
model learning
Feature extraction
standard language
Genetic algorithms
Neural networks
neural network
performance
Spoken Language
Machine Learning
regression analysis
Processing
Language

Keywords

  • Extreme learning machine
  • Language identification
  • Optimised genetic algorithm

ASJC Scopus subject areas

  • Software
  • Language and Linguistics
  • Human-Computer Interaction
  • Linguistics and Language
  • Computer Vision and Pattern Recognition

Cite this

Spoken language identification based on optimised genetic algorithm–extreme learning machine approach. / Albadr, Musatafa Abbas Abbood; Tiun, Sabrina; Ayob, Masri; AL-Dhief, Fahad Taha.

In: International Journal of Speech Technology, 01.01.2019.

Research output: Contribution to journalArticle

@article{1baafc59554d4462a14175157be036de,
title = "Spoken language identification based on optimised genetic algorithm–extreme learning machine approach",
abstract = "The determination and classification of a recognized spoken language based on certain contents and datasets is known as the process of language identification (LID). The common process in carrying out LID entails the mandatory processing of data which enables the extraction of the necessary features for the process. The extraction involves a mature process whereby the development of the standard LID features have been conducted much earlier by means of a mel-frequency cepstral coefficients, shifted delta cepstral, Gaussian mixture model and i-vector-based framework. Despite that, improvement or rather optimisation still needs to be done on the learning process based on the extracted features so as to obtain all the knowledge embedded within them. The classification and regression analysis can benefit tremendously from the use of the extreme learning machine (ELM) which is a particularly effective and useful learning model for training a single-hidden layer neural network. However, owing to the randomly selected weights embedded in the input’s hidden layers, the model’s learning process is rendered to be ineffective or not optimised in its entirety. In this study, the ELM is employed as the learning model for LID due to the standard feature extraction. In addition, this study proposes a new optimised genetic algorithm (OGA) with three different selection criteria (i.e., roulette wheel, K-tournament and random) to select the appropriate initial weights and biases of the input hidden layer of the ELM, thereby minimising the classification error and improving the general performance of the ELM for LID. Results show the excellent performance of the proposed OGA–ELM with three different selection criteria, namely, roulette wheel, K-tournament and random, with the highest accuracies of 99.50{\%}, 100{\%} and 99.38{\%}, respectively.",
keywords = "Extreme learning machine, Language identification, Optimised genetic algorithm",
author = "Albadr, {Musatafa Abbas Abbood} and Sabrina Tiun and Masri Ayob and AL-Dhief, {Fahad Taha}",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s10772-019-09621-w",
language = "English",
journal = "International Journal of Speech Technology",
issn = "1381-2416",
publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Spoken language identification based on optimised genetic algorithm–extreme learning machine approach

AU - Albadr, Musatafa Abbas Abbood

AU - Tiun, Sabrina

AU - Ayob, Masri

AU - AL-Dhief, Fahad Taha

PY - 2019/1/1

Y1 - 2019/1/1

N2 - The determination and classification of a recognized spoken language based on certain contents and datasets is known as the process of language identification (LID). The common process in carrying out LID entails the mandatory processing of data which enables the extraction of the necessary features for the process. The extraction involves a mature process whereby the development of the standard LID features have been conducted much earlier by means of a mel-frequency cepstral coefficients, shifted delta cepstral, Gaussian mixture model and i-vector-based framework. Despite that, improvement or rather optimisation still needs to be done on the learning process based on the extracted features so as to obtain all the knowledge embedded within them. The classification and regression analysis can benefit tremendously from the use of the extreme learning machine (ELM) which is a particularly effective and useful learning model for training a single-hidden layer neural network. However, owing to the randomly selected weights embedded in the input’s hidden layers, the model’s learning process is rendered to be ineffective or not optimised in its entirety. In this study, the ELM is employed as the learning model for LID due to the standard feature extraction. In addition, this study proposes a new optimised genetic algorithm (OGA) with three different selection criteria (i.e., roulette wheel, K-tournament and random) to select the appropriate initial weights and biases of the input hidden layer of the ELM, thereby minimising the classification error and improving the general performance of the ELM for LID. Results show the excellent performance of the proposed OGA–ELM with three different selection criteria, namely, roulette wheel, K-tournament and random, with the highest accuracies of 99.50%, 100% and 99.38%, respectively.

AB - The determination and classification of a recognized spoken language based on certain contents and datasets is known as the process of language identification (LID). The common process in carrying out LID entails the mandatory processing of data which enables the extraction of the necessary features for the process. The extraction involves a mature process whereby the development of the standard LID features have been conducted much earlier by means of a mel-frequency cepstral coefficients, shifted delta cepstral, Gaussian mixture model and i-vector-based framework. Despite that, improvement or rather optimisation still needs to be done on the learning process based on the extracted features so as to obtain all the knowledge embedded within them. The classification and regression analysis can benefit tremendously from the use of the extreme learning machine (ELM) which is a particularly effective and useful learning model for training a single-hidden layer neural network. However, owing to the randomly selected weights embedded in the input’s hidden layers, the model’s learning process is rendered to be ineffective or not optimised in its entirety. In this study, the ELM is employed as the learning model for LID due to the standard feature extraction. In addition, this study proposes a new optimised genetic algorithm (OGA) with three different selection criteria (i.e., roulette wheel, K-tournament and random) to select the appropriate initial weights and biases of the input hidden layer of the ELM, thereby minimising the classification error and improving the general performance of the ELM for LID. Results show the excellent performance of the proposed OGA–ELM with three different selection criteria, namely, roulette wheel, K-tournament and random, with the highest accuracies of 99.50%, 100% and 99.38%, respectively.

KW - Extreme learning machine

KW - Language identification

KW - Optimised genetic algorithm

UR - http://www.scopus.com/inward/record.url?scp=85069050746&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069050746&partnerID=8YFLogxK

U2 - 10.1007/s10772-019-09621-w

DO - 10.1007/s10772-019-09621-w

M3 - Article

AN - SCOPUS:85069050746

JO - International Journal of Speech Technology

JF - International Journal of Speech Technology

SN - 1381-2416

ER -