Proceedings of the First International Conference on Advanced Scientific Innovation in Science, Engineering and Technology, ICASISET 2020, 16-17 May 2020, Chennai, India

Research Article

Audio Codec Using Legendre Functions for Simplified Digital Telephony

  • @INPROCEEDINGS{10.4108/eai.16-5-2020.2304205,
        author={Sabapathy  Ananthi and Ananthanarayanan  Arun and Sivakannan  Subramani and Krishnasamy  Padmanabhan},
        title={Audio Codec Using Legendre Functions for Simplified Digital Telephony},
        proceedings={Proceedings of the First  International Conference on Advanced Scientific Innovation in Science, Engineering and Technology, ICASISET 2020, 16-17 May 2020, Chennai, India},
        publisher={EAI},
        proceedings_a={ICASISET},
        year={2021},
        month={1},
        keywords={legendre polynomial celp audio codec’s speech processing lpc},
        doi={10.4108/eai.16-5-2020.2304205}
    }
    
  • Sabapathy Ananthi
    Ananthanarayanan Arun
    Sivakannan Subramani
    Krishnasamy Padmanabhan
    Year: 2021
    Audio Codec Using Legendre Functions for Simplified Digital Telephony
    ICASISET
    EAI
    DOI: 10.4108/eai.16-5-2020.2304205
Sabapathy Ananthi1,*, Ananthanarayanan Arun2, Sivakannan Subramani3, Krishnasamy Padmanabhan3
  • 1: Dept. of Network Systems and IT, University of Madras, Chennai, India
  • 2: Dept. of ECE, MVJ College of Engineering, Bangalore, India
  • 3: Department of Bigdata Analytics, St.Joseph’s College, Bangalore
*Contact email: ananthipradeep84@gmail.com

Abstract

Speech signal is based on the more complex anatomy of the mouth, throat, and nasal bone. Such speech signals are controlled by digital signal processing for encoding and by compressing the same speech. Many speech signal processing methods and standards have been developed by many researchers for audio codec’s such as CELP. Advanced mathematical transformations such as Fourier, Lagrange, Hamilton, Legendre were developed for such applications. Signal processing using its equations through certain algorithms and methods to develop signals for various novel applications. Based on laryngeal pulses and oral cavity filter patterns, these techniques can be used to encode speech patterns more efficiently without losing signal clarity. In this paper a novel method of encoding speech with much simpler computations has been evolved. The famous spherical harmonic Legendre polynomials have been the key to the technique. Good compression, decent sound quality and above all, simplicity are the claimed advantages of the method. A new approach, using the spherical harmonic oscillator functions of Legendre is introduced here for this application. Thus, this paper describes this alternate scheme for speech compression and coding. Its usefulness lies in the simplicity relative to the more involved CELP techniques currently in use.