Volume 8 Issue 2 - March 2017

  • 1. Vowel recognition for cochlear implant device testing

    Authors : Gautami Pujare

    Pages : 1-9

    DOI : http://dx.doi.org/10.21172/1.82.001

    Keywords : Keywords- Pitch, Timbre, Cepstral coefficient

    Abstract :

    Abstract- Human listeners are better able to identify vowels on the basis of timbre characterization. Timbre is the attribute that distinguishes sounds of equal pitch, loudness and duration. It contributes to our perception and discrimination of different vowels and consonants in speech, instruments in music and environmental sounds. The Mel Frequency Cepstral Coefficient vectors are commonly used in audio analysis. They are described as timbral features as they model the short-time spectral characteristics of the signal onto a psychoacoustic frequency scale.In this work an audio signal analysis approach based on Mel Frequency Cepstral Coefficients, for vowel recognition is proposed. Mel Frequency Cepstral analysis is utilized as a feature extraction technique. Mel Frequency Cepstral coefficients extracted for vowel phonemes are used as template and matched with audio input signal. To recognize vowels, Adaptive Network based Fuzzy Inference System (ANFIS) is used as a feature matching technique. In simulation results, vowel is recognized correctly, which can be useful in tuning cochlear implant device in medical field. It also can be used for capturing the timbre information of music signals.

    Citing this Journal Article :

    Gautami Pujare, "Vowel recognition for cochlear implant device testing", https://www.ijltet.org/journal_details.php?id=911&j_id=3650, Volume 8 Issue 2 - March 2017, 1-9, #ijltetorg