ML vs. MAP PARAMETER ESTIMATION OF LINEAR DYNAMIC SYSTEMS FOR ACOUSTIC-TO-ARTICULATORY INVERSION: A COMPARATIVE STUDY

Date

2010-08-27

Author

Özbek Arslan, Işıl
Demirekler, Mübeccel

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

164
views

0
downloads

This work proposes a maximum a posteriori (MAP) based parameter learning algorithm for acoustic-to-articulatory inversion. Inversion method is based on single global linear dynamic system (GLDS) representation of acoustic and articulatory data. MAP based learning algorithm considers a prior distribution for the parameter set as well as the likelihood of the training data. Therefore in this paper, we investigate the selection of prior distributions with hyperparameters for GLDS to improve the performance of articulatory inversion. The performance of the proposed learning algorithm and comparison of it with the maximum likelihood (ML) based learning method are examined on an extensive set of examples. These results show that the performance of the articulatory inversion method based on GLDS is significantly improved via MAP based learning algorithm.

Subject Keywords

Maximum likelihood estimation , Hidden Markov models, Speech , Correlation , Acoustics , Learning systems , Trajectory

URI

https://hdl.handle.net/11511/55931

Conference Name

18th European Signal Processing Conference (EUSIPCO)

Collections

Department of Basic English, Conference / Seminar

Suggestions

OpenMETU
Core

A 2-D unsteady Navier-Stokes solution method with overlapping/overset moving grids Tuncer, İsmail Hakkı (1996-01-01) A simple, robust numerical algorithm to localize intergrid boundary points and to interpolate unsteady solution variables across 2-D, overset/overlapping, structured computational grids is presented. Overset/ overlapping grids are allowed to move in time relative to each other. The intergrid boundary points are localized in terms of three grid points on the donor grid by a directional search algorithm. The final parameters of the search algorithm give the interpolation weights at the interpolation point. Th...
Principal Coordinate Clustering SEKMEN, ali; ALDROUBİ, Akram; HAMM, Keaton; Koku, Ahmet Buğra (2017-12-14) This paper introduces a clustering algorithm, called principal coordinate clustering. It takes in a similarity matrix SW of a data matrix W and computes the singular value decomposition of SW to determine the principal coordinates to convert the clustering problem to a simpler domain. It is a relative of spectral clustering, however, principal coordinate clustering is easier to interpret, and gives a clear understanding of why it performs well. In a fashion, this gives intuition behind why spectral clusteri...
2-D Navier-Stokes solution method with overset moving grids Tuncer, İsmail Hakkı (1996-01-01) A simple, robust numerical algorithm to localize moving boundary points and to interpolate unsteady solution variables across 2-D, arbitrarily overset computational grids is presented. Overset grids are allowed to move in time relative to each other. The intergrid boundary points are localized in terms of three grid points on the donor grid by a directional search algorithm. The parameters of the search algorithm give the interpolation weights at the localized boundary point. The method is independent of nu...
ML Estimation of Process Noise Variance in Dynamic Systems Axelsson, Patrik; Orguner, Umut; Gustafsson, Fredrik; Norrlöf, Mikael (2011-08-28) The performance of a non-linear filter hinges in the end on the accuracy of the assumed non-linear model of the process. In particular, the process noise covariance Q is hard to get by physical modeling and dedicated system identification experiments. We propose a variant of the expectation maximization (EM) algorithm which iteratively estimates the unobserved state sequence and Q based on the observations of the process. The extended Kalman smoother (EKS) is the instrument to find the unobserved state sequ...
Image segmentation by fusion of low level and domain specific information via Markov Random Fields Karadag, Ozge Oztimur; Yarman Vural, Fatoş Tunay (2014-09-01) We propose a new segmentation method by fusing a set of top-down and bottom-up segmentation maps under the Markov Random Fields (MRF) framework. The bottom-up segmentation maps are obtained by varying the parameters of an unsupervised segmentation method, such as Mean Shift. The top-down segmentation maps are constructed from some priori information, called domain specific information (DSI), received from a domain expert in the form of general properties about the image dataset. The properties are then used...

Citation Formats

I. Özbek Arslan and M. Demirekler, “ML vs. MAP PARAMETER ESTIMATION OF LINEAR DYNAMIC SYSTEMS FOR ACOUSTIC-TO-ARTICULATORY INVERSION: A COMPARATIVE STUDY,” presented at the 18th European Signal Processing Conference (EUSIPCO), Aalborg, DENMARK, 2010, Accessed: 00, 2020. [Online]. Available: https://hdl.handle.net/11511/55931.