Estimating Age in Short Utterances Based on Multi-Class Classification Approach

作     者：Ameer A.Badr Alia K.Abdul-Hassan 

作者机构：College of Managerial and Financial SciencesImam Ja’afar Al-Sadiq UniversitySalahaddinIraq Department of Computer ScienceUniversity of TechnologyBaghdadIraq 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2021年第68卷第8期

页      面：1713-1729页

核心收录：

学科分类：07[理学] 0701[理学-数学] 070101[理学-基础数学] 

主　　题：Speaker age estimation XGBoost statistical functionals Quantile normalization LDA TIMIT dataset 

摘      要：Age estimation in short speech utterances finds many applications in daily life like human-robot interaction,custom call routing,targeted marketing,user-profiling,*** the comprehensive studies carried out to extract descriptive features,the estimation errors(***)are still *** this study,an automatic system is proposed to estimate age in short speech utterances without depending on the text as well as the ***,four groups of features are extracted from each utterance frame using hybrid techniques and *** that,10 statistical functionals are measured for each extracted feature ***,the extracted feature dimensions are normalized and reduced using the Quantile method and the Linear Discriminant Analysis(LDA)method,***,the speaker’s age is estimated based on a multi-class classification approach by using the Extreme Gradient Boosting(XGBoost)*** have been carried out on the TIMIT dataset to measure the performance of the proposed *** Mean Absolute Error(MAE)of the suggested system is 4.68 years,and 4.98 years,the Root Mean Square Error(RMSE)is 8.05 and 6.97,respectively,for female and male *** results show a clear relative improvement in terms of MAE up to 28%and 10%for female and male speakers,respectively,in comparison to related works that utilized the TIMIT dataset.