Matrix form of the LSP–HS method
DOI:
https://doi.org/10.20535/RADAP.2014.56.13-24Keywords:
encoding shape of the spectral envelope, linear spectral parameters, LSP, linear spectral pairs, linear spectral frequencies, LSF, linear prediction coefficients, LPCAbstract
Introduction. The coding theory of the forms of spectral envelope of speech signals of the using linear spectral parameters of the highest splitting (LSP–HS) is regarded. Advantages of using of LSP–HS parameters compare to the classic LSP parameters are specified.
Main part. It is demonstrated that the direct and inverse transform of LSP–HS method can be regarded as a certain matrices transform from the LPC coefficients. Method of determining the matrix elements of direct and inverse transformation method LSP–HS is described. It is shown that the matrix of direct and inverse transformation of method LSP–HS can be calculated for arbitrary prediction order. The results of determination of the elements of specified matrices for 10th prediction order are given. Basis vectors of the direct and inverse transformation matrices of LSP–HS method are plotted on the charts. The formula of estimates of the spectral envelope of the synthesized speech signal in space LSP–HS is written.
Conclusions. It is proved that the method of LSP–HS can be used in matrix form. The matrix form of LSP–HS method makes it possible to employ standard mathematical tools to describe the shape of the spectral envelope of speech signal not only in space LPC, but in the space of LSP–HS, which can not be done in the space of classical (first phase splitting) LSP. The matrix form of LSP–HS method allows to draw alternative structure of filters of analysis and synthesis of speech not with classic LPC parameters, but with the coefficients of LSP–HS space, and to reveal the physical meaning of the transition from LPC to the LSP in algorithms of speech coding.
References
Перелік посилань:
Павлов О.И. Прямое П–преобразование в линейном предсказании речи // Известия вузов. Радиоэлектроника. – 2000. – Т. 43, № 12. – С. 53–66.
Павлов О.И. Упрощение реализации метода линейных спектральных пар (частот) в линейном предсказании речи // Труды 3–й Международной конференции “Цифровая обработка сигналов и ее применение”, Т.3, с. 128–132, Москва, 2000.
Павлов О.И. Быстрый алгоритм и графическое представление прямого преобразования в методе линейных спектральных частот высшего порядка // Труды 3–й Международной конференции “Цифровая обработка сигналов и ее применение”, Т. 3, с. 132–136, Москва, 2000.
Павлов О.И. Алгоритм быстрого прямого П–преобразования и особенности его математического аппарата // Известия вузов. Радиоэлектроника. – 2001. – Т. 44, № 2. – С. 61–73.
Павлов О.И. Обратное П–преобразование в линейном предсказании речи // Известия вузов. Радиоэлектроника. – 2001. – Т. 44, № 1. – С. 61-73.
Павлов О.И. Алгоритм быстрого обратного П–преобразования // Радиоэлектроника. – 2001. – Т. 44, № 8.– С. 67 – 77.
Павлов О.И. Свойства линейных спектральных частот высших порядков // Радиотехника: Всеукр. межвед. науч.-техн. сб. – 2001. – Вып. 117. – С. 62 – 64.
Павлов О.И. Межкадровая интерполяция спектральной огибающей речевого сигнала в пространстве линейных спектральных частот наивысшей регрессии // Известия вузов. Радиоэлектроника. – 2008. – Т. 51, № 4.– С. 56 – 69.
Павлов О.І. Оцінка ефективності кодування спектральної обвідної мовленнєвих сигналів в просторах лінійних спектральних параметрів найвищої регресії методом кла-стерного аналізу / О.І. Павлов, П.А. Стасевич, Г.М. Тертичний // Праці ІХ всеукр. міжн. конф. “Оброблення сигналів і зображень та розпізнавання образів, УкрОбраз’2008”, 3–7 лист. 2008р. – Київ: МННЦ ІТіС НАН та МОН України, УАсОІРО, 2008. – С. 189 – 192. Режим доступу – http://uasoiro.kibermova.com/files/Zbirnyk/2008/5/p_50.pdf.
Павлов О.І., Герасименко К.В., Аполонов Є.В. Прогнозування спектральної обвідної мовленнєвих сигналів в просторі лінійних спектральних параметрів найвищо-го розщеплення // Праці 11–ої Всеукраїнської міжнародної конференції “Оброблення сигналів і зображень та розпізнавання образів”, УкрОбраз’2012, 15–19 жовтня 2012 р. – Київ : МННЦ ІТіС НАН та МОН України, УАсОІРО, 2012. – С. 137–144.
References:
Pavlov O. I. (2000) Priamoe P–preobrazovanie v lineinom predskazanii rechi [Direct P–transformation in the linear prediction of speech]. Radioelectronics and Communications Systems, vol. 43, no. 12, pp. 53–66.
Pavlov O. I. (2000) Uproshchenie realizatsii metoda lineinikh spektralnikh par (chastot) v lineinom predskazanii rechi [Simplifying the implementation of the linear spectral pairs (frequencies) in the linear prediction speech] Trudy 3 Mezhdunarodnoi konferentsii “Tsyfrovaia obrabotka sihnalov i ee primenenie” [Proc. 3th Int. Conf. “ Digital Signal Processing and Applications”], Vol. 3, Moskow, pp. 128–132.
Pavlov O. I. Bystryi alhorytm i hraficheskoe predstavlenie priamoho preobrazovaniia v metode lineinikh spektralnikh chastot vyssheho poriadka [Fast algorithm and graphical representation of the direct transformations in method of line spectral frequencies higher order]. Trudy 3 Mezhdunarodnoi konferentsii “Tsyfrovaia obrabotka sihnalov i ee primenenie” [Proc. 3th Int. Conf. “ Digital Signal Processing and Applications”], Vol. 3, Moskow, pp. 132 – 136.
Pavlov O. I. (2001) Alhoritm bystroho priamoho P–preobrazovaniia i osobennosti eho matematicheskoho apparata [Algorithm for fast direct P-transformation and features of its mathematical apparatus]. Radioelectronics and Communications Systems, vol. 44, no. 2, pp. 61–73.
Pavlov O. I. (2001) Obratnoe P–preobrazovanie v lineinom predskazanii rechi [Inverse P-transformation in linear prediction of speech]. Radioelectronics and Communications Systems, vol. 44, no. 1, pp. 61 – 73.
Pavlov O. I. (2001) Alhoritm bystroho obratnoho P–preobrazovaniia [Algorithm for fast inverse P-transformation]. Radioelectronics and Communications Systems, vol. 44, no. 8, pp. 67 – 77.
Pavlov O. I. (2001) Svoistva lineinykh spektralnykh chastot vysshikh poriadkov [Properties of line spectral frequencies higher orders]. Radiotekhnika: Vseukr. mezhved. nauch.-tekhn. sb. Vol. 117. pp. 62 – 64.
Pavlov O. I. (2008) Inter-frame interpolation of the spectral envelope of the speech signal in the space of linear spectral frequencies of the highest regression Radioelectronics and Communications Systems. vol. 51, no. 4, pp. 56 – 69. doi: 10.3103/S0735272708040067
Pavlov O. I., Stasevych P. A. and Tertychnyy H. M. (2008) Estimation of speech signals spectrum envelope encoding efficiency in linear spectrum parameters spaces of the highe st regression with a cluster analysis method. Obroblennia syhnaliv i zobrazhen ta rozpiznavannia obraziv, UkrObraz2008 [Proc. Of 9-th Int. Conf “Signal/Image Processing And Pattern Recognition, UkrOBRAZ'2008”], Kyiv, pp. 189 – 192. (in Ukrainian) Available at: http://uasoiro.kibermova.com/files/Zbirnyk/2008/5/p_50.pdf
Pavlov O. I., Herasymenko K. V. and Apollonov Ye. V. (2012) Speech signals spectral envelope prediction in the linear spectral parameter space of highest splitting. Obroblennia syhnaliv i zobrazhen ta rozpiznavannia obraziv, UkrObraz2008 [Proc. Of 11-th Int. Conf “Signal/Image Processing And Pattern Recognition, UkrOBRAZ'2012”], Kyiv, pp. 137–144. (in Ukrainian)
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
