Speech AnalysisSynthesis Based on a Sinusoidal Representation-91H

High Quality and Low Complexity Speech Analysis/Synthesis Based on Sinusoidal Representation

Speech Analysis/Synthesis Based on a Sinusoidal Representation.

Also some speech coding methods have been applied to speech synthesis, such as Linear Predictive Coding and Sinusoidal Modeling. Actually, the first speech synthesizer, VODER, was developed from the speech coding system VOCODER (Klatt 1987, Schroeder 1993). Linear Prediction has been used for several decades, but with the basic method the quality has been quite poor. However, with some modifications, such as Warped Linear Prediction (WLP), considerable achievements have been reported (Karjalainen et al. 1998). Warped filtering takes advantage of hearing properties, so it is perhaps useful in all source-filter based synthesis methods. Sinusoidal models have also been applied to speech synthesis for about a decade. Like PSOLA methods, the sinusoidal modeling is best suited for periodic signals, but the representation of unvoiced speech is difficult. However, the sinusoidal methods have been found useful with singing voice synthesis (Macon 1996).

Speech analysis/synthesis based on a sinusoidal representation.

Speech analysis/synthesis based on a sinusoidal representation

Description :This thesis develops a series of programs that implement the sinusoidal representation model for speech and sound waveform analysis and synthesis. This sinusoidal representation model can also be used...

Speech Processing Based on a Sinusoidal Model Using a sinusoidal model of speech, an analysis/synthesis technique has been ..

In this paper, we discuss several improvements in sinusoidal coding. First, we present an analysis method which estimates windowed sinusoids to represent a segment of an input speech or audio signal. The window used in the analysis is the same magnitude-complementary window as is used in overlap-add synthesis, which makes analysis consistent with synthesis. It is hown how the overlapping nature of segments can be accounted for in the sinusoidal estimation. Second, we present techniques for optimization of sinusoidal parameters based on the squared difference between the input signal and reconstruction. Efficient methods for computation are also discussed. Experimental results verify that our procedures provide a significant improvement in reconstruction accuracy.

Conclusions. In this paper an analysis/synthesis technique based on a sinusoidal representation was presented that has proven to be very appropriate for …