Robust Speaker Recognition Systems with Adaptive Filter Algorithms in Real Time Under Noisy Conditions

  • Hema Kumar PentapatiEmail author
  • Madhu Tenneti
Conference paper
Part of the Learning and Analytics in Intelligent Systems book series (LAIS, volume 4)


At present, Speaker recognition systems are being widely used in speaker detection, authentication and authorization to perform secure transactions in various personal, commercial and industrial applications. As speaker recognition in noisy environments is becoming increasingly difficult, a novel system is developed using MFCC and Vector Quantization. The training data is collected from 15 native speakers. LMS, NLMS and RLS adaptive filters are used to reduce the noise in the speech signal. The performance of all the speaker recognition systems is rated by calculating the Equal Error Rate (ERR) and Euclidian distance.


Mel Frequency Cepstral Coefficients (MFCC) Least mean square Normalised LMS Recursive Least Squares Euclidian distance Equal Error Rate Vector Quantization 


  1. 1.
    Nijhawan G, Soni MK (2014) Speaker recognition using MFCC and vector quantisation. Int. J. Recent Trends Eng Technol 11(1):211Google Scholar
  2. 2.
    Hemakumar P, Srinivas V, Madhu T (2016) Improved dynamic speaker recognition system using NLMS adaptive filter. Int J Comput Appl (0975–8887) 148(10):39–43Google Scholar
  3. 3.
    Kekre HB, Bharadi VA, Sawant AR (2012) Speaker recognition using vector quantization by MFCC and KMCG clustering algorithm. In: 2012 international conference on communication, information & computing technology (ICCICT), 19–20 October 2012Google Scholar
  4. 4.
    Srinivasa Kumar Ch, Mallikarjuna Rao P (2011) Design of an automatic speaker recognition system using MFCC, vector quantization and LBG algorithm. Int J Comput Sci Eng (IJCSE) 3(8). ISSN 0975-3397Google Scholar
  5. 5.
    Abdul Samad S, Hussin A, Anuar Ishak K (2015) Improving hybrid speaker verification in noisy environment using least mean square adaptive filters. In: The 5th International Conference on Information and Communication Technology for The Muslim World (ICT4M), 17–18 November 2014. ISBN 978-1-4799-6242-6Google Scholar
  6. 6.
    Poddar A, Sahidullah Md, Saha G (2015) Performance comparison of speaker recognition systems in presence of duration variability. 978–1–4673–6540–6/15/$31.00 @ 2015 IEEEGoogle Scholar
  7. 7.
    Dhiman J, Ahmad S, Gulia K (2013) Comparison between adaptive filter algorithms (LMS, NLMS and RLS). Int J Sci Eng Technol Res (IJSETR) 2(5):1100–1103Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Department of ECESwarnandhra Institute of Engineering and TechnologyNarasapur, West Godavari DistrictIndia

Personalised recommendations