Sound dimensions and formants
DOI:
https://doi.org/10.36505/ExLing-2020/11/0023/000438Keywords:
Three-dimensional depiction, Chebyshev Transform, formantAbstract
Every physical event that can be observed can be measured and described, including sounds. This paper discusses computer algorithms that were developed to depict vowels and speech sounds in their three dimensions: frequency, energy, and time. Each vowel has a separate distinguishable shape based on its dimensions. Two-dimensional vowel plots can be more accurately represented in three-dimensional plots. Algorithms using the Chebyshev Transform were written and vowel speech signals were converted to accurate numerical data sets that were examined and then plotted. Comparisons of vowels can be made, based on their sonic shape. This algorithm also used the Singular Value Decomposition (SVD) to measure, vowel formants giving clear formant regions with the frequency regions identified on the y-axis plots.
References
Johnson, K. 2012. Acoustics and Auditory Phonetics. MA, Wiley-Blackwell Malden.
Boyd, J. 2001. Chebyshev and Fourier Spectral Methods. Mineola, NY, Dover Publications, Inc.
Gold, B., Morgan, N. 2000. Signal Processing and Perception of Speech and Music. NY, John Wiley & Sons.
https://www.youtube.com/user/EugeneKhutoryansky. January 25, 2019.
Trefethen, L. 2019., Spectral Methods in MATLAB, SIAM, 2000.
Downloads
Published
Issue
Section
License
Articles are published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are properly credited.
