Homomorphic analysis is a well-known method for the separation of non-linearly combined signals More particularly, the use of complex cepstrum for source-tract deconvolution has been discussed in various articles However there exists no study which proposes a glottal flow estimation methodology based on cepstrum and reports effective results In this paper, we show that complex cepstrum can be effectively used for glottal flow estimation by separating the causal and anticausal components of a windowed speech signal as done by the Zeros of the Z-Transform (ZZT) decomposition Based on exactly the same principles presented for ZZT decomposition, windowing should be applied such that the windowed speech signals exhibit mixed-phase characteristics which conform the speech production model that the anticausal component is mainly due to the glottal flow open phase The advantage of the complex cepstrum-based approach compared to the ZZT decomposition is its much higher speed Index Terms: Speech Analysis, Homomorphic Processing, Glottal Source Estimation

/pdf/complex-cepstrum-based-decomposition-of-speech-for-glottal-3oikk0vll2.pdf

Complex Cepstrum-based Decomposition of Speech for Glottal Source Estimation

Complex cepstrum is known in the literature for linearly separating causal and anticausal components. Relying on advances achieved by the Zeros of the Z-Transform (ZZT) technique, we here investigate the possibility of using complex cepstrum for glottal flow estimation on a large-scale database. Via a systematic study of the windowing effects on the deconvolution quality, we show that the complex cepstrum causal-anticausal decomposition can be effectively used for glottal flow estimation when specific windowing criteria are met. It is also shown that this complex cepstral decomposition gives similar glottal estimates as obtained with the ZZT method. However, as complex cepstrum uses FFT operations instead of requiring the factoring of high-degree polynomials, the method benefits from a much higher speed. Finally in our tests on a large corpus of real expressive speech, we show that the proposed method has the potential to be used for voice quality analysis.

Causal-Anticausal Decomposition of Speech using Complex Cepstrum for Glottal Source Estimation

https://hal.archives-ouvertes.fr/hal-00746109/document

Causal-anticausal decomposition of speech using complex cepstrum for glottal source estimation

The extraction of the third dimension from remote sensing data is a well known technique. Since in a number of countries aerial images and laser scanner data are unavailable, expensive or classified, stereoscopic high-resolution optical satellite images provide a viable alternative for generating digital surface and digital terrain models. Especially the automatic extraction of highly accurate 3D surface models in urban areas is still a very complicated task due to occlusions, large differences in height and the variety of objects and surface material. In this paper an analysis and a visual and quantitative comparison of three different matching algorithms for generating urban DSMs based on very high-resolution satellite images is presented. The three algorithms are least squares matching (LSM) in a region growing fashion, dynamic programming (DP) and semiglobal matching (SGM). The characteristics of the three algorithms as applied to four different Ikonos stereo pairs with a ground sampling distance of 1 m are shown. The following results were obtained: visually, in the LSM results the shape of the buildings is considerably smoothed. While in the DP results the building shape is sharper, only little detail is visible on the building roofs, and streaking along the epipolar lines causes problems. With SGM more details can be extracted and the results visually have the best quality. Based on reference data for the different test sites, the standard deviation of the building heights determined by LSM and DP is in the range of one pixel or slightly better, while it is in the range of half a pixel for SCM.

Comparison of Matching Algorithms for DSM Generation in Urban Areas from Ikonos Imagery

Complex cepstrum for statistical parametric speech synthesis

Traditionally, a very simple model for short-time homomorphic analysis has been used. It is shown that there is no theoretical justification for applying this model to voiced speech and that the model is of limited value for improving cepstral deconvolution procedures. Consequently, a more elaborate model is introduced in which the influence of window length is approximated and the spectral sampling inherent in voiced speech is explicitly represented. As a result, this new model shows that the vocal tract contribution to the complex cepstrum is repeated at every multiple of the pitch quefrency (n p ) and is multiplied by a double sinclike distortion (D(n)). It is shown that in order to achieve deconvolution with a low-time gating system, a cepstral lifter of length n p /2 should be used (instead of the usual length "less than n p "). Furthermore, the lifter should compensate for the distortion D(n). Unfortunately, the accuracy of straightforward homomorphic deconvolution approximations is limited by aliasing distortion which results from the repeated nature of the vocal tract contribution. Nevertheless, reasonable deconvolution approximations are obtained.

A new model for the short-time complex cepstrum of voiced speech

It is shown that the need for an intermediate estimate of the spectral envelope can be avoided in the determination of group delay characteristics from the short-time cepstrum, provided that the analysis parameters are carefully chosen.

O. Steenhaut

Papers

A new model for the short-time complex cepstrum of voiced speech

Comments on "Pole-zero analysis of voiced speech using group delay characteristics"