Multi-speaker articulatory reconstruction based on an eigen articulatory HMM

We present a multi-speaker articulatory reconstruction method based on speaker-independent articulatory features and speaker-dependent features. These features are separated by using a multi-speaker articulatory database. This separation method consists of normalizing palate positions among the spea...

Full description

Saved in:
Bibliographic Details
Main Authors: Hiroya, S., Mochida, T.
Format: Conference Proceeding
Language:English
Subjects:
Acoustic measurements
Covariance matrix
Hidden Markov models
Laboratories
Loudspeakers
Maximum likelihood linear regression
Reconstruction algorithms
Sampling methods
Spatial databases
Speech
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a multi-speaker articulatory reconstruction method based on speaker-independent articulatory features and speaker-dependent features. These features are separated by using a multi-speaker articulatory database. This separation method consists of normalizing palate positions among the speakers and separating multi-speaker articulatory data into a speaker-independent eigen articulatory HMM and a speaker-adaptive matrix by using speaker-adaptive training (SAT). With the proposed method, the average RMS errors of the measured and reconstructed articulatory parameters were 1.35 mm. This result shows that the proposed method makes it possible to control speaker idiosyncrasies in the articulatory parameter domain.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2005.1415262