Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization

Stephen Nyakuti Otenyi; Livingstone Ngoo; Henry Kiragu

doi:doi:10.11648/j.ijiis.20241305.11

Research Article |

| Peer-Reviewed

Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization

Stephen Nyakuti Otenyi^*

, Livingstone Ngoo

, Henry Kiragu

Published in International Journal of Intelligent Information Systems (Volume 13, Issue 5)

Received: 3 July 2024 Accepted: 5 August 2024 Published: 10 October 2024

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

Though there are advancements in speaker recognition technology, available systems often fail to correctly recognize speakers especially in noisy environments. The use of Mel-frequency cepstral coefficients (MFCC) has been improved using Convolutional Neural Networks (CNN) yet difficulties in achieving high accuracies still exists. Hybrid algorithms combining MFCC and Region-based Convolutional Neural Networks (RCNN) have been found to be promising. In this research features from speech signals were extracted for speaker recognition, to denoise the signals, design and develop a DFT-based denoising system using spectrum subtraction and to develop a speaker recognition method for the Verbatim Transcription using MFCC. The DFT was used to transform the sampled audio signal waveform into a frequency-domain signal. RCNN was used to model the characteristics of speakers based on their voice samples, and to classify them into different categories or identities. The novelty of the research was that it used MFCC integrated with RCNN and optimized with Host-Cuckoo Optimization (HCO) algorithm. HCO algorithm is capable of further weight optimization through the process of generating fit cuckoos for best weights. It also captured the temporal dependencies and long-term information. The system was tested and validated on audio recordings from different personalities from the National Assembly of Kenya. The results were compared with the actual identity of the speakers to confirm accuracy. The performance of the proposed approach was compared with two other existing speaker recognition the traditional approaches being MFCC-CNN and Linear Predictive Coefficients (LPC)-CNN. The comparison was based the Equal Error Rate (EER), False Rejection Rate (FRR), False Match Rate (FMR), and True Match Rate (TMR). Results show that the proposed algorithm outperformed the others in maintaining a lowest EER, FMR, FRR and highest TMR.

Published in	International Journal of Intelligent Information Systems (Volume 13, Issue 5)
DOI	10.11648/j.ijiis.20241305.11
Page(s)	94-108
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Speaker Recognition, Verbatim Transcription, Spectrum Subtraction, Mel-Frequency Cepstral Coefficients (MFCC), Linear Predictive Cepstral Coefficients, HCO Algorithm

References

[1]	R. M. Hanifa, K. Isa, and S. Mohamad, “A review on speaker recognition: Technology and challenges,” Comput. Electr. Eng., vol. 90, p. 107005, 2021. https://doi.org/10.1016/j.compeleceng.2021.107005
[2]	M. Jakubec, E. Lieskovska, and R. Jarina, “An Overview of Automatic Speaker Recognition in Adverse Acoustic Environment,” presented at the 2020 18th International Conference on Emerging eLearning Technologies and Applications (ICETA), IEEE, 2020, pp. 211–218. https://doi.org/10.1109/iceta51985.2020.9379245
[3]	D. Hershcovich et al., “Challenges and strategies in cross-cultural NLP,” ArXiv Prepr. ArXiv220310020, 2022. https://doi.org/10.18653/v1/2022.acl-long.482
[4]	V. N. Ngoni, “English–Bukusu Automatic Machine Translation for Digital Services Inclusion in E-governance,” 2022.
[5]	N. O. Ogechi, “On language rights in Kenya,” Nord. J. Afr. Stud., vol. 12, no. 3, pp. 19–19, 2003.
[6]	S. S. Tirumala, S. R. Shahamiri, A. S. Garhwal, and R. Wang, “Speaker identification features extraction methods: A systematic review,” Expert Syst. Appl., vol. 90, pp. 250–271, 2017. https://doi.org/10.1016/j.eswa.2017.08.015
[7]	National Assembly of Kenya, “Standing Orders.” National Assembly of Kenya, 2013. [Online]. Available: http://www.parliament.go.ke/sites/default/files/2022-08/National%20Assembly%20Standing%20Orders%20-%206th%20Edition,%202022_0.pdf
[8]	R. Jahangir et al., “Text-independent speaker identification through feature fusion and deep neural network,” IEEE Access, vol. 8, pp. 32187–32202, 2020. https://doi.org/10.1109/access.2020.2973541
[9]	G. Sharma, K. Umapathy, and S. Krishnan, “Trends in audio signal feature extraction methods,” Appl. Acoust., vol. 158, p. 107020, 2020. https://doi.org/10.1016/j.apacoust.2019.107020
[10]	J. V. E. López, “Adaptation of Speaker and Speech Recognition Methods for theAutomatic Screening of Speech Disorders Using Machine Learning,” 2023. https://doi.org/10.14232/phd.11491
[11]	M. M. Kabir, M. F. Mridha, J. Shin, I. Jahan, and A. Q. Ohi, “A survey of speaker recognition: Fundamental theories, recognition methods and opportunities,” IEEE Access, vol. 9, pp. 79236–79263, 2021. https://doi.org/10.1109/access.2021.3084299
[12]	S. J. Jainar, P. L. Sale, and B. Nagaraja, “VAD, feature extraction and modelling techniques for speaker recognition: a review,” Int. J. Signal Imaging Syst. Eng., vol. 12, no. 1–2, pp. 1–18, 2020. https://doi.org/10.1504/ijsise.2020.10036128
[13]	S. K. Sarangi and G. Saha, “Improved speech-signal based frequency warping scale for cepstral feature in robust speaker verification system,” J. Signal Process. Syst., vol. 92, pp. 679–692, 2020. https://doi.org/10.1007/s11265-020-01517-2
[14]	S. Bharadwaj and P. B. Acharjee, “Analysis of Prosodic features for the degree of emotions of an Assamese Emotional Speech,” presented at the 2020 4th International Conference on Electronics, Communication and Aerospace Technology (ICECA), IEEE, 2020, pp. 1441–1452. https://doi.org/10.1109/iceca49313.2020.9297453
[15]	C. Quam and S. C. Creel, “Impacts of acoustic‐phonetic variability on perceptual development for spoken language: A review,” Wiley Interdiscip. Rev. Cogn. Sci., vol. 12, no. 5, p. e1558, 2021. https://doi.org/10.1002/wcs.1558
[16]	S. Ali, S. Tanweer, S. S. Khalid, and N. Rao, “Mel frequency cepstral coefficient: a review,” ICIDSSD, 2020. https://doi.org/10.4108/eai.27-2-2020.2303173
[17]	S. Pangaonkar and A. Panat, “A Review of Various Techniques Related to Feature Extraction and Classification for Speech Signal Analysis,” presented at the ICDSMLA 2019: Proceedings of the 1st International Conference on Data Science, Machine Learning and Applications, Springer, 2020, pp. 534–549. https://doi.org/10.1007/978-981-15-1420-3_57
[18]	B. A. Aicha and F. Kacem, “Conventional Machine Learning and Feature Engineering for Vocal Fold Precancerous Lesions Detection Using Acoustic Features,” Circuits Syst. Signal Process., pp. 1–33, 2023. https://doi.org/10.1007/s00034-023-02551-8
[19]	K. Jagadeeshwar, T. Sreenivasarao, P. Pulicherla, K. Satyanarayana, K. M. Lakshmi, and P. M. Kumar, “ASERNet: Automatic speech emotion recognition system using MFCC-based LPC approach with deep learning CNN,” Int. J. Model. Simul. Sci. Comput., vol. 14, no. 04, p. 2341029, 2023. https://doi.org/10.1142/s1793962323410295
[20]	M. Ramashini, P. E. Abas, K. Mohanchandra, and L. C. De Silva, “Robust cepstral feature for bird sound classification,” Int. J. Electr. Comput. Eng., vol. 12, no. 2, p. 1477, 2022. https://doi.org/10.11591/ijece.v12i2.pp1477-1487
[21]	G. Vanderreydt and K. Demuynck, “A Novel Channel estimate for noise robust speech recognition,” Comput. Speech Lang., p. 101598, 2023. https://doi.org/10.1016/j.csl.2023.101598
[22]	H. Hermansky and N. Morgan, “RASTA processing of speech,” IEEE Trans. Speech Audio Process., vol. 2, no. 4, pp. 578–589, 1994. https://doi.org/10.1109/89.326616
[23]	N. N. Alrouqi, “Additive Noise Subtraction for Environmental Noise in Speech Recognition,” 2021.
[24]	S. Alharbi et al., “Automatic speech recognition: Systematic literature review,” IEEE Access, vol. 9, pp. 131858–131876, 2021. https://doi.org/10.1109/access.2021.3112535
[25]	Wen-Jie Song, Chen Chen, Tian-Yang Sun, and Wei Wang, “A Robust Equalization Feature for Language Recognition,” J. Inf. Sci. Eng., vol. 36, no. 3, pp. 561–576, May 2020, https://doi.org/10.6688/JISE.202005_36(3).0006
[26]	G. Manikandan and S. Abirami, “Feature Selection Is Important: State-of-the-Art Methods and Application Domains of Feature Selection on High-Dimensional Data,” in Applications in Ubiquitous Computing, R. Kumar and S. Paiva, Eds., in EAI/Springer Innovations in Communication and Computing. Cham: Springer International Publishing, 2021, pp. 177–196. https://doi.org/10.1007/978-3-030-35280-6_9
[27]	O. Ghahabi, P. Safari, and J. Hernando, “Deep Learning in Speaker Recognition,” in Development and Analysis of Deep Learning Architectures, W. Pedrycz and S.-M. Chen, Eds., in Studies in Computational Intelligence, Cham: Springer International Publishing, 2020, pp. 145–169. https://doi.org/10.1007/978-3-030-31764-5_6
[28]	Y. Wang, L. Zheng, Y. Gao, and S. Li, “Vibration Signal Extraction Based on FFT and Least Square Method,” IEEE Access, vol. 8, pp. 224092–224107, 2020, https://doi.org/10.1109/ACCESS.2020.3044149
[29]	J. Agrawal, M. Gupta, and H. Garg, “A review on speech separation in cocktail party environment: challenges and approaches,” Multimed. Tools Appl., vol. 82, no. 20, pp. 31035–31067, Aug. 2023, https://doi.org/10.1007/s11042-023-14649-x
[30]	P. Golik, “Data-driven deep modeling and training for automatic speech recognition,” 2020.
[31]	R. Haeb-Umbach, J. Heymann, L. Drude, S. Watanabe, M. Delcroix, and T. Nakatani, “Far-Field Automatic Speech Recognition,” Proc. IEEE, vol. 109, no. 2, pp. 124–148, Feb. 2021, https://doi.org/10.1109/JPROC.2020.3018668
[32]	A. Lauraitis, R. Maskeliūnas, R. Damaševičius, and T. Krilavičius, “Detection of Speech Impairments Using Cepstrum, Auditory Spectrogram and Wavelet Time Scattering Domain Features,” IEEE Access, vol. 8, pp. 96162–96172, 2020, https://doi.org/10.1109/ACCESS.2020.2995737
[33]	S. A. El-Moneim et al., “Speaker recognition based on pre-processing approaches,” Int. J. Speech Technol., vol. 23, no. 2, pp. 435–442, Jun. 2020, https://doi.org/10.1007/s10772-019-09659-w
[34]	N. Chen and S. Fu, “Uncertainty quantification of nonlinear Lagrangian data assimilation using linear stochastic forecast models,” Phys. Nonlinear Phenom., vol. 452, p. 133784, Oct. 2023, https://doi.org/10.1016/j.physd.2023.133784
[35]	A. P. Fellows, M. T. L. Casford, and P. B. Davies, “Spectral Analysis and Deconvolution of the Amide I Band of Proteins Presenting with High-Frequency Noise and Baseline Shifts,” Appl. Spectrosc., vol. 74, no. 5, pp. 597–615, May 2020, https://doi.org/10.1177/0003702819898536
[36]	N. Saleem, J. Gao, M. I. Khattak, H. T. Rauf, S. Kadry, and M. Shafi, “DeepResGRU: Residual gated recurrent neural network-augmented Kalman filtering for speech enhancement and recognition,” Knowl.-Based Syst., vol. 238, p. 107914, Feb. 2022, https://doi.org/10.1016/j.knosys.2021.107914
[37]	P. Bansal, S. A. Imam, and R. Bharti, “Speaker recognition using MFCC, shifted MFCC with vector quantization and fuzzy,” presented at the 2015 International Conference on Soft Computing Techniques and Implementations (ICSCTI), IEEE, 2015, pp. 41–44. https://doi.org/10.1109/icscti.2015.7489535
[38]	L.-M. Dogariu, J. Benesty, C. Paleologu, and S. Ciochină, “An Insightful Overview of the Wiener Filter for System Identification,” Appl. Sci., vol. 11, no. 17, Art. no. 17, Jan. 2021, https://doi.org/10.3390/app11177774
[39]	Y. Zouhir, M. Zarka, and K. Ouni, “Power Normalized Gammachirp Cepstral (PNGC) coefficients-based approach for robust speaker recognition,” Appl. Acoust., vol. 205, p. 109272, Mar. 2023, https://doi.org/10.1016/j.apacoust.2023.109272
[40]	A. Ahmed, Y. Serrestou, K. Raoof, and J.-F. Diouris, “Empirical Mode Decomposition-Based Feature Extraction for Environmental Sound Classification,” Sensors, vol. 22, no. 20, Art. no. 20, Jan. 2022, https://doi.org/10.3390/s22207717
[41]	Z. Bai and X.-L. Zhang, “Speaker recognition based on deep learning: An overview,” Neural Netw., vol. 140, pp. 65–99, 2021. https://doi.org/10.1016/j.neunet.2021.03.004
[42]	Gaurav, S. Bhardwaj, and R. Agarwal, “An efficient speaker identification framework based on Mask R-CNN classifier parameter optimized using hosted cuckoo optimization (HCO),” J. Ambient Intell. Humaniz. Comput., vol. 14, no. 10, pp. 13613–13625, Oct. 2023, https://doi.org/10.1007/s12652-022-03828-7
[43]	Z. Touati-Hamad and M. R. Laouar, “Enhancing Education Decision-Making with Deep Learning for Arabic Spoken Digit Recognition,” p. 4321405 Bytes, 2023, https://doi.org/10.6084/M9.FIGSHARE.24902382.V3
[44]	W.-C. Lin and C. Busso, “Chunk-Level Speech Emotion Recognition: A General Framework of Sequence-to-One Dynamic Temporal Modeling,” IEEE Trans. Affect. Comput., vol. 14, no. 2, pp. 1215–1227, Apr. 2023, https://doi.org/10.1109/TAFFC.2021.3083821
[45]	S. Agarwal, J. O. D. Terrail, and F. Jurie, “Recent Advances in Object Detection in the Age of Deep Convolutional Neural Networks.” arXiv, Aug. 20, 2019. https://doi.org/10.48550/arXiv.1809.03193
[46]	C. Zhang, Z. Yang, X. He, and L. Deng, “Multimodal Intelligence: Representation Learning, Information Fusion, and Applications,” IEEE J. Sel. Top. Signal Process., vol. 14, no. 3, pp. 478–493, Mar. 2020, https://doi.org/10.1109/JSTSP.2020.2987728
[47]	S. Hourri, N. S. Nikolov, and J. Kharroubi, “Convolutional neural network vectors for speaker recognition,” Int. J. Speech Technol., vol. 24, no. 2, pp. 389–400, Jun. 2021, https://doi.org/10.1007/s10772-021-09795-2
[48]	G. Hu, Z. Zhang, A. Armaou, and Z. Yan, “Robust extended Kalman filter based state estimation for nonlinear dynamic processes with measurements corrupted by gross errors,” J. Taiwan Inst. Chem. Eng., vol. 106, pp. 20–33, Jan. 2020, https://doi.org/10.1016/j.jtice.2019.10.015
[49]	O. Deshpande, K. Solanki, S. P. Suribhatla, S. Zaveri, and L. Ghodasara, “Simulating the DFT Algorithm for Audio Processing,” ArXiv Prepr. ArXiv210502820, 2021.
[50]	M. Awais, Md. T. Bin Iqbal, and S.-H. Bae, “Revisiting Internal Covariate Shift for Batch Normalization,” IEEE Trans. Neural Netw. Learn. Syst., vol. 32, no. 11, pp. 5082–5092, Nov. 2021, https://doi.org/10.1109/TNNLS.2020.3026784
[51]	H. Moayed and E. G. Mansoori, “Improving Regularization in Deep Neural Networks by Co-adaptation Trace Detection,” Neural Process. Lett., vol. 55, no. 6, pp. 7985–7997, Dec. 2023, https://doi.org/10.1007/s11063-023-11293-2
[52]	R. Gadagkar, “How to Design Experiments in Animal Behaviour,” Resonance, vol. 25, no. 10, pp. 1419–1455, Oct. 2020, https://doi.org/10.1007/s12045-020-1061-4

Cite This Article

Plain Text BibTeX RIS

APA Style

Otenyi, S. N., Ngoo, L., Kiragu, H. (2024). Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization. International Journal of Intelligent Information Systems, 13(5), 94-108. https://doi.org/10.11648/j.ijiis.20241305.11

Copy | Download

ACS Style

Otenyi, S. N.; Ngoo, L.; Kiragu, H. Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization. Int. J. Intell. Inf. Syst. 2024, 13(5), 94-108. doi: 10.11648/j.ijiis.20241305.11

Copy | Download

AMA Style

Otenyi SN, Ngoo L, Kiragu H. Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization. Int J Intell Inf Syst. 2024;13(5):94-108. doi: 10.11648/j.ijiis.20241305.11

Copy | Download

@article{10.11648/j.ijiis.20241305.11,
  author = {Stephen Nyakuti Otenyi and Livingstone Ngoo and Henry Kiragu},
  title = {Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization
},
  journal = {International Journal of Intelligent Information Systems},
  volume = {13},
  number = {5},
  pages = {94-108},
  doi = {10.11648/j.ijiis.20241305.11},
  url = {https://doi.org/10.11648/j.ijiis.20241305.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijiis.20241305.11},
  abstract = {Though there are advancements in speaker recognition technology, available systems often fail to correctly recognize speakers especially in noisy environments. The use of Mel-frequency cepstral coefficients (MFCC) has been improved using Convolutional Neural Networks (CNN) yet difficulties in achieving high accuracies still exists. Hybrid algorithms combining MFCC and Region-based Convolutional Neural Networks (RCNN) have been found to be promising. In this research features from speech signals were extracted for speaker recognition, to denoise the signals, design and develop a DFT-based denoising system using spectrum subtraction and to develop a speaker recognition method for the Verbatim Transcription using MFCC. The DFT was used to transform the sampled audio signal waveform into a frequency-domain signal. RCNN was used to model the characteristics of speakers based on their voice samples, and to classify them into different categories or identities. The novelty of the research was that it used MFCC integrated with RCNN and optimized with Host-Cuckoo Optimization (HCO) algorithm. HCO algorithm is capable of further weight optimization through the process of generating fit cuckoos for best weights. It also captured the temporal dependencies and long-term information. The system was tested and validated on audio recordings from different personalities from the National Assembly of Kenya. The results were compared with the actual identity of the speakers to confirm accuracy. The performance of the proposed approach was compared with two other existing speaker recognition the traditional approaches being MFCC-CNN and Linear Predictive Coefficients (LPC)-CNN. The comparison was based the Equal Error Rate (EER), False Rejection Rate (FRR), False Match Rate (FMR), and True Match Rate (TMR). Results show that the proposed algorithm outperformed the others in maintaining a lowest EER, FMR, FRR and highest TMR.
},
 year = {2024}
}

Copy | Download

TY - JOUR
T1 - Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization

AU - Stephen Nyakuti Otenyi
AU - Livingstone Ngoo
AU - Henry Kiragu
Y1 - 2024/10/10
PY - 2024
N1 - https://doi.org/10.11648/j.ijiis.20241305.11
DO - 10.11648/j.ijiis.20241305.11
T2 - International Journal of Intelligent Information Systems
JF - International Journal of Intelligent Information Systems
JO - International Journal of Intelligent Information Systems
SP - 94
EP - 108
PB - Science Publishing Group
SN - 2328-7683
UR - https://doi.org/10.11648/j.ijiis.20241305.11
AB - Though there are advancements in speaker recognition technology, available systems often fail to correctly recognize speakers especially in noisy environments. The use of Mel-frequency cepstral coefficients (MFCC) has been improved using Convolutional Neural Networks (CNN) yet difficulties in achieving high accuracies still exists. Hybrid algorithms combining MFCC and Region-based Convolutional Neural Networks (RCNN) have been found to be promising. In this research features from speech signals were extracted for speaker recognition, to denoise the signals, design and develop a DFT-based denoising system using spectrum subtraction and to develop a speaker recognition method for the Verbatim Transcription using MFCC. The DFT was used to transform the sampled audio signal waveform into a frequency-domain signal. RCNN was used to model the characteristics of speakers based on their voice samples, and to classify them into different categories or identities. The novelty of the research was that it used MFCC integrated with RCNN and optimized with Host-Cuckoo Optimization (HCO) algorithm. HCO algorithm is capable of further weight optimization through the process of generating fit cuckoos for best weights. It also captured the temporal dependencies and long-term information. The system was tested and validated on audio recordings from different personalities from the National Assembly of Kenya. The results were compared with the actual identity of the speakers to confirm accuracy. The performance of the proposed approach was compared with two other existing speaker recognition the traditional approaches being MFCC-CNN and Linear Predictive Coefficients (LPC)-CNN. The comparison was based the Equal Error Rate (EER), False Rejection Rate (FRR), False Match Rate (FMR), and True Match Rate (TMR). Results show that the proposed algorithm outperformed the others in maintaining a lowest EER, FMR, FRR and highest TMR.

VL - 13
IS - 5
ER -

Copy | Download

Author Information

Stephen Nyakuti Otenyi

Electrical and Telecommunication Engineering Department, Multimedia University of Kenya, Nairobi, Kenya

Contact Email

http://orcid.org/0009-0004-0012-5863
Livingstone Ngoo

Electrical and Telecommunication Engineering Department, Multimedia University of Kenya, Nairobi, Kenya

Contact Email

http://orcid.org/0000-0003-0009-9701
Henry Kiragu

Electrical and Telecommunication Engineering Department, Multimedia University of Kenya, Nairobi, Kenya

Contact Email

http://orcid.org/0009-0008-9976-2713

Download PDF

Sections

Plain Text BibTeX RIS

APA Style

Otenyi, S. N., Ngoo, L., Kiragu, H. (2024). Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization. International Journal of Intelligent Information Systems, 13(5), 94-108. https://doi.org/10.11648/j.ijiis.20241305.11

Copy | Download

ACS Style

Otenyi, S. N.; Ngoo, L.; Kiragu, H. Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization. Int. J. Intell. Inf. Syst. 2024, 13(5), 94-108. doi: 10.11648/j.ijiis.20241305.11

Copy | Download

AMA Style

Otenyi SN, Ngoo L, Kiragu H. Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization. Int J Intell Inf Syst. 2024;13(5):94-108. doi: 10.11648/j.ijiis.20241305.11

Copy | Download

@article{10.11648/j.ijiis.20241305.11,
  author = {Stephen Nyakuti Otenyi and Livingstone Ngoo and Henry Kiragu},
  title = {Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization
},
  journal = {International Journal of Intelligent Information Systems},
  volume = {13},
  number = {5},
  pages = {94-108},
  doi = {10.11648/j.ijiis.20241305.11},
  url = {https://doi.org/10.11648/j.ijiis.20241305.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijiis.20241305.11},
  abstract = {Though there are advancements in speaker recognition technology, available systems often fail to correctly recognize speakers especially in noisy environments. The use of Mel-frequency cepstral coefficients (MFCC) has been improved using Convolutional Neural Networks (CNN) yet difficulties in achieving high accuracies still exists. Hybrid algorithms combining MFCC and Region-based Convolutional Neural Networks (RCNN) have been found to be promising. In this research features from speech signals were extracted for speaker recognition, to denoise the signals, design and develop a DFT-based denoising system using spectrum subtraction and to develop a speaker recognition method for the Verbatim Transcription using MFCC. The DFT was used to transform the sampled audio signal waveform into a frequency-domain signal. RCNN was used to model the characteristics of speakers based on their voice samples, and to classify them into different categories or identities. The novelty of the research was that it used MFCC integrated with RCNN and optimized with Host-Cuckoo Optimization (HCO) algorithm. HCO algorithm is capable of further weight optimization through the process of generating fit cuckoos for best weights. It also captured the temporal dependencies and long-term information. The system was tested and validated on audio recordings from different personalities from the National Assembly of Kenya. The results were compared with the actual identity of the speakers to confirm accuracy. The performance of the proposed approach was compared with two other existing speaker recognition the traditional approaches being MFCC-CNN and Linear Predictive Coefficients (LPC)-CNN. The comparison was based the Equal Error Rate (EER), False Rejection Rate (FRR), False Match Rate (FMR), and True Match Rate (TMR). Results show that the proposed algorithm outperformed the others in maintaining a lowest EER, FMR, FRR and highest TMR.
},
 year = {2024}
}

Copy | Download

TY - JOUR
T1 - Speaker Recognition System Using Hybrid of MFCC and RCNN with HCO Algorithm Optimization

VL - 13
IS - 5
ER -

Copy | Download