350 rub
Journal Science Intensive Technologies №5 for 2012 г.
Article in number:
Measurement of the fundamental tone of a speech signal on the basis of its autocorrelation function
Authors:
A.S. Kolokolov, I.A. Lubinski, A.Yu. Mestcherakov
Abstract:
The methods of measurement of the fundamental tone based on the autocorrelation function of the natural and the center-clipped speech signal are considered. In these methods the fundamental frequency is determined by as the inverse coordinate the main peak of the autocorrelation function of the signal. Proposed and investigated a modified autocorrelation method for measuring the fundamental tone, based on the procedure used by clipping the autocorrelation function of the speech signal. Demonstrated its advantages over methods based on autocorrelation of natural and centrally clipped speech signal, consisting of the lower sensitivity of the measurement results of the resonances of the vocal tract and a greater resistance to the additive impulsive noise.
Pages: 26-29
References
  1. Ли У.А. Применение просодического анализа для распознавания речи. В кн.: Методы автоматического распознавания речи / под ред. У.А. Ли. М.: Мир. 1983.
  2. Бондарос Ю.Г., Колоколов А.С., Костюк А.И. Исследование речевых сигналов в условиях кабины летательного аппарата // Вестник компьютерных и информационных технологий. 2008. №4. С. 2-10.
  3. Фант Г. Акустическая теория речеобразования. М.: Наука. 1964.
  4. Fant G. Speech Acoustics and Phonetics // Kluwer Academic Publishers. 2004. 
  5. Sondhi M.M. New methods of pitch extractio // IEEE Trans. Audio and Electroacoust. 1968. V. AU-16. № 2. P. 262-266.
  6. Рабинер Л.Р., Шафер Р.В. Цифровая обработка речевых сигналов. М.: Радиоисвязь. 1981.
  7. Boersma P. Accurate short-term analysis of the fundamental frequency and harmonic to nose ratio of sampled sound // Inast. PhoneticSci., Univer. Amsterdam. 1993. Proc. 19. P. 97-110.
  8. Колоколов А.С. Измерение основного тона речевого сигнала // Автоматика и телемеханика. 2003. № 8. C. 122-134.