Improving acoustic-to-articulatory inversion by using hypercube codebooks
نویسندگان
چکیده
This paper presents an articulatory codebook construction method which gives a good space coverage with a limited number of points. We represent the articulatory space by hypercubes. Therefore the articulatory space is decomposed into regions represented by few number of points. We describe our interpolation method to retrieve points from the hypercube and the inversion method based on the same interpolation method. The advantage of the codebook and the inversion method is its robustness with respect to articulatoryto-acoustic mapping non-linearity problems.
منابع مشابه
Design of hypercube codebooks for the acoustic-to-articulatory inversion respecting the non-linearities of the articulatory-to-acoustic mapping
This paper presents a new method to generate a codebook that best represents the mapping between articulatory and acoustic domains. Existing codebook generation methods are relatively inefficient because of the non-linearities between the two domains. In our method the articulatory space is considered as composed of several hypercubes where the mapping is linear. The advantages of such a repres...
متن کاملExploring the null space of the acoustic-to- articulatory inversion using a hypercube codebook
Our acoustic to articulatory inversion method exploits an original codebook representing the articulatory space by hypercubes. The articulatory space is decomposed into regions where the articulatory-to-acoustic mapping is linear. Each region is represented by a hypercube. The inversion procedure retrieves articulatory vectors corresponding to an acoustic entry from the hypercube codebook. The ...
متن کاملModeling the articulatory space using a hypercube codebook for acoustic-to-articulatory inversion.
Acoustic-to-articulatory inversion is a difficult problem mainly because of the nonlinearity between the articulatory and acoustic spaces and the nonuniqueness of this relationship. To resolve this problem, we have developed an inversion method that provides a complete description of the possible solutions without excessive constraints and retrieves realistic temporal dynamics of the vocal trac...
متن کاملIntroduction of constraints in an acoustic-to-articulatory inversion method based on a hypercubic articulatory table
Our acoustic to articulatory inversion method exploits an original articulatory table structured in the form of a hypercube hierarchy. The articulatory space is decomposed into regions where the articulatory-to-acoustic mapping is linear. Each region is represented by a hypercube. The inversion procedure retrieves articulatory vectors corresponding to an acoustic entry from the hypercube table....
متن کاملA non-linear filtering approach to stochastic training of the articulatory-acoustic mapping using the EM algorithm
Current techniques for training representations of the articulatory-acoustic mapping from data rely on arti cial simulations to provide codebooks of articulatory and acoustic measurements, which are then modelled by simple functional approximations. This paper outlines a stochastic framework for adapting an arti cial model to real speech from acoustic measurements alone, using the EM algorithm....
متن کامل