Pub Date : 2024-02-01DOI: 10.33889/ijmems.2024.9.1.008
Asmita Nirmal, Deepak Jayaswal, P. Kachare
A speaker verification system models individual speakers using different speech features to improve their robustness. However, redundant features degrade the system's performance. This paper presents Statistically Significant Duration-Independent Mel frequency Cepstral Coefficients (SSDI-MFCC) features with the Extreme Gradient Boost classifier for improving the noise-robustness of speaker models. Eight statistical descriptors are used to generate signal duration-independent features, and a statistically significant feature subset is obtained using a t-test. A redeveloped Librispeech database by adding noises from the AURORA database to simulate real-world test conditions for speaker verification is used for evaluation. The SSDI-MFCC is compared with Principal Component Analysis (PCA) and Genetic Algorithm (GA). The comparative results showed average equal error rate improvements by 4.93 % and 3.48 % with the SSDI-MFCC than GA-MFCC and PCA-MFCC in clean and noisy conditions, respectively. A significant reduction in verification time is observed using SSDI-MFCC than the complete feature set.
{"title":"Statistically Significant Duration-Independent-based Noise-Robust Speaker Verification","authors":"Asmita Nirmal, Deepak Jayaswal, P. Kachare","doi":"10.33889/ijmems.2024.9.1.008","DOIUrl":"https://doi.org/10.33889/ijmems.2024.9.1.008","url":null,"abstract":"A speaker verification system models individual speakers using different speech features to improve their robustness. However, redundant features degrade the system's performance. This paper presents Statistically Significant Duration-Independent Mel frequency Cepstral Coefficients (SSDI-MFCC) features with the Extreme Gradient Boost classifier for improving the noise-robustness of speaker models. Eight statistical descriptors are used to generate signal duration-independent features, and a statistically significant feature subset is obtained using a t-test. A redeveloped Librispeech database by adding noises from the AURORA database to simulate real-world test conditions for speaker verification is used for evaluation. The SSDI-MFCC is compared with Principal Component Analysis (PCA) and Genetic Algorithm (GA). The comparative results showed average equal error rate improvements by 4.93 % and 3.48 % with the SSDI-MFCC than GA-MFCC and PCA-MFCC in clean and noisy conditions, respectively. A significant reduction in verification time is observed using SSDI-MFCC than the complete feature set.","PeriodicalId":517298,"journal":{"name":"International Journal of Mathematical, Engineering and Management Sciences","volume":"105 3","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139897671","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-01DOI: 10.33889/ijmems.2024.9.1.002
Vishnu Uppalakkal, Venkatesh Ambati, Rajesh Nair
This study investigates the effectiveness of the Firefly Optimizer (FFA), Grey Wolf Optimizer (GWO), and Moth Flame Optimizer (MFO) metaheuristic algorithms in estimating the kinetic parameters of a single-step coal pyrolysis model. By examining the effects of the algorithmic configuration, the initial parameter estimates, and the search space size on the efficacy and efficiency of the optimization run, the research seeks to encourage the qualified engineering application of these algorithms in the field of pyrolysis modeling. Four critical analyses were conducted: convergence efficiency, robustness and repeatability, parameter tuning, and performance on noisy data. MFO and GWO had comparable fitness scores of 1.05×10-4 and 1.04×10-4 respectively in the optimisation run analysis, while FireFly Algorithm (FFA) fell behind with a score of 1.09×10-4. Regarding the calculation time, FFA showed better results than other optimizers with an execution time of 113.75 seconds. MFO showed initial promise in convergence analysis with speedy convergence, whereas GWO progressively enhanced its solutions. Additionally, GWO was shown to be the most dependable algorithm with the lowest values for average fitness score and execution time at 1.07×10-4 and 38.86 seconds. The combined values of standard deviation in fitness value and execution time for GWO were 1.07×10-6 and 0.35 indicating its robustness towards initial parameters. Similar to this, investigations on repeatability emphasized the reliability of the GWO method. Further, the parameter tuning assessments supported the balanced performance of GWO, and the studies of noise handling discovered GWO to be the most robust to noisy data. Overall, GWO is recommended as a one-stop average solution for the general engineered application; however, algorithm choice hinges on the specific requirement.
{"title":"Performance Assessment of Metaheuristic Algorithms: Firefly, Grey Wolf, and Moth Flame in Coal Pyrolysis Kinetic Parameter Estimation","authors":"Vishnu Uppalakkal, Venkatesh Ambati, Rajesh Nair","doi":"10.33889/ijmems.2024.9.1.002","DOIUrl":"https://doi.org/10.33889/ijmems.2024.9.1.002","url":null,"abstract":"This study investigates the effectiveness of the Firefly Optimizer (FFA), Grey Wolf Optimizer (GWO), and Moth Flame Optimizer (MFO) metaheuristic algorithms in estimating the kinetic parameters of a single-step coal pyrolysis model. By examining the effects of the algorithmic configuration, the initial parameter estimates, and the search space size on the efficacy and efficiency of the optimization run, the research seeks to encourage the qualified engineering application of these algorithms in the field of pyrolysis modeling. Four critical analyses were conducted: convergence efficiency, robustness and repeatability, parameter tuning, and performance on noisy data. MFO and GWO had comparable fitness scores of 1.05×10-4 and 1.04×10-4 respectively in the optimisation run analysis, while FireFly Algorithm (FFA) fell behind with a score of 1.09×10-4. Regarding the calculation time, FFA showed better results than other optimizers with an execution time of 113.75 seconds. MFO showed initial promise in convergence analysis with speedy convergence, whereas GWO progressively enhanced its solutions. Additionally, GWO was shown to be the most dependable algorithm with the lowest values for average fitness score and execution time at 1.07×10-4 and 38.86 seconds. The combined values of standard deviation in fitness value and execution time for GWO were 1.07×10-6 and 0.35 indicating its robustness towards initial parameters. Similar to this, investigations on repeatability emphasized the reliability of the GWO method. Further, the parameter tuning assessments supported the balanced performance of GWO, and the studies of noise handling discovered GWO to be the most robust to noisy data. Overall, GWO is recommended as a one-stop average solution for the general engineered application; however, algorithm choice hinges on the specific requirement.","PeriodicalId":517298,"journal":{"name":"International Journal of Mathematical, Engineering and Management Sciences","volume":"15 10","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139897312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：