Comparison of methods for determining speech voicing based on tests performed on paired consonants and continuous speech

Jan Malucha; Milan Sigmund

doi:10.2478/jee-2022-0049

Abstract

Voicing is an important phonetic characteristic of speech. Each phoneme belongs to a group of either voiced or unvoiced sounds. We investigated and compared the performance of five algorithms widely used to estimate speech voicing. All algorithms were implemented in Matlab and tested on both short consonants and continuous speech. Phonetically paired consonants (voiced vs unvoiced) and parts of read speech from audio books were used in the experiments. The tuned harmonics-to-noise ratio method gave the best results in both situations, ie for consonants and continuous speech. Using this method, the overall voicing of Czech, Polish, Hungarian and English was investigated. Hungarian speech showed the highest proportion of voiced parts, approx. 75 %. In other languages, the proportion of voiced parts was around 70 %.

References

[1] International Phonetic Association, Handbook of the International Phonetic Association A Guide to the Use of the International Phonetic Alphabet, Cambridge University Press, 1999.
Search in Google Scholar Back to article
[2] https://www.internationalphoneticassociation.org/IPAcharts/inter_chart_2018/IPA_2018.html.
Search in Google Scholar Back to article
[3] R. Sussex and P. Cubberley, The Slavic Languages, Cambridge University Press, 2006.10.1017/CBO9780511486807
Search in Google Scholar Back to article
[4] D. Odden, Introducing Phonology, New York, Cambridge University Press, 2005.10.1017/CBO9780511808869
Search in Google Scholar Back to article
[5] M. J. Ball, S. J. Howard, and K. Miller, “Revisions to the ExtIPA,”, Journal of the International Phonetic Association, pp. 156–164.
Search in Google Scholar Back to article
[6] M. Sigmund, A. Prokes, and Z. Brabec, “Statistical Analysis of Glottal Pulses in Speech under Psychological Stress,”, 16th European Signal Processing Conference (EUSIPCO), pp. 1–5, 2008.
Search in Google Scholar Back to article
[7] M. M. Sondhi, “Measurement of the Glottal Waveform,”, The Journal of the Acoustical Society of America, no. 1, pp. 228–232, 1975.10.1121/1.3804291110287
Search in Google Scholar Back to article
[8] R. L. Miller, “Nature of the Vocal Cord Wave,”, The Journal of the Acoustical Society of America, no. 6, pp. 667–677, 1959.10.1121/1.1907771
Search in Google Scholar Back to article
[9] J. Psutka, L. Müller, J. Matoušek, and V. Radová, Mluvíme s počítačem česky, Prague: Academia, 2006, (in Czech).
Search in Google Scholar Back to article
[10] R. G. Bachu, S. Kopparthi, B. Adapa, and B. D. Barkana, “Voiced/unvoiced Decision for Speech Signals Based on Zero Crossing Rate and Energy,”, Advanced Techniques in Computing Sciences and Software Engineering, Dordrecht: Springer, pp. 279–282, 2010.10.1007/978-90-481-3660-5_47
Search in Google Scholar Back to article
[11] J. Heranová, Harmonicita jako možný indiktor hranic mezi segmenty v češtině, Praha, Univerzita Karlova, Filozofická fakulta, Fonetickýústav, 2010, (in Czech).
Search in Google Scholar Back to article
[12] L. R. Rabiner and R. W. Schafer, Theory and Applications of Digital Speech Processing, London: Prentice Hall, 2011.
Search in Google Scholar Back to article
[13] H. Misra, S. Ikbal, H. Bourlard, and H. Hermansky, “Spectral Entropy Based Feature for Robust ASR,”, Int. Conference on Acoustics, Speech, and Signal Processing, pp. I193–I196, 2004.
Search in Google Scholar Back to article
[14] P. J. Murphy and O. O. Akande, “Noise Estimation in Voice Signals Using Short-Term Cepstral Analysis,”, The Journal of the Acoustical Society of America, no. 3, pp. 1679–1690, 2007.
Search in Google Scholar Back to article
[15] Y. Wei, Y. Zeng, and C. Li, “Single-Channel Speech Enhancement Based on Subband Spectral Entropy,”, Journal of the Audio Engineering Society, no. 3, pp. 100–113, 2018.10.17743/jaes.2018.0003
Search in Google Scholar Back to article
[16] M. Sigmund, “Statistical Analysis of Fundamental Frequency Based Features in Speech Under Stress”, Information Technology and Control, no. 3, pp. 286–291, 2013.10.5755/j01.itc.42.3.3895
Search in Google Scholar Back to article
[17] P. Boersma and D. Weenink, “Praat: doing phonetics by computer”, http://www.praat.org, 2022.
Search in Google Scholar Back to article
[18] M. Stanek and M. Sigmund, “Psychological Stress Detection in Speech Using Return-to-Opening Phase Ratios in Glottis,”, Elektronika ir Elektrotechnika, no. 5, pp. 59–63, 2015.10.5755/j01.eie.21.5.13336
Search in Google Scholar Back to article
[19] J. Pribil, A. Pribilova, and J. Matousek, “Evaluation of Synthetic Speech Quality by Statistical Analysis of Voiced and Un-voiced Part Durations,”, Int. Conference on Telecommunications and Signal Processing (TSP), pp. 1–4, 2018.10.1109/TSP.2018.8441352
Search in Google Scholar Back to article
[20] M. Sigmund and T. Dostal, “Analysis of Emotional Stress in Speech,”, International Conference on Artificial Intelligence and Applications, Innsbruck, pp. 317–322, 2004.
Search in Google Scholar Back to article
[21] S. Sondhi, R. Vijay, M. Khan, and A. K. Salhan, “Voice Analysis for Detection of Deception,”, Int. Conference on Knowledge, Information and Creativity Support Systems, pp. 1–6, 2016.10.1109/KICSS.2016.7951455
Search in Google Scholar Back to article
[22] A. Bayestehtashk, M. Asgari, I. Shafran, and J. McNames, “Fully Automated Assessment of the Severity of Parkinson’s Disease from Speech,”, Computer Speech and Language, no. 1, pp. 172–185, 2015.10.1016/j.csl.2013.12.001422205425382935
Search in Google Scholar Back to article
[23] M. L. B. Pulido, J. B. A. Hernández, M. A. F. Ballester, C. M. T. González, J. Mekyska, and Z. Smékal, “Alzheimer’s Disease and Automatic Speech Analysis: A Review,”, Expert Systems with Applications, pp. 1–19, 2020.10.1016/j.eswa.2020.113213
Search in Google Scholar Back to article
[24] P. Zelinka and M. Sigmund, “Hierarchical Classification Tree Modeling of Nonstationary Noise for Robust Speech Recognition”, Information Technology and Control, no. 3, pp. 202–210, 2010.
Search in Google Scholar Back to article
[25] P. Zelinka and M. Sigmund, “Automatic Vocal E ort Detection for Reliable Speech Recognition”, Int. Workshop on Machine Learning for Signal Processing, Kittila, pp. 349–354, 2010.10.1109/MLSP.2010.5589174
Search in Google Scholar Back to article

Comparison of methods for determining speech voicing based on tests performed on paired consonants and continuous speech

Abstract

Paradigm

My account