Data processing: speech signal processing – linguistics – language – Speech signal processing – Recognition
Reexamination Certificate
2003-06-20
2008-11-18
Dorvil, Richemond (Department: 2626)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Recognition
C704S242000, C704S255000, C704S256100, C704S256400, C704S256700
Reexamination Certificate
active
07454336
ABSTRACT:
A system and method that facilitate modeling unobserved speech dynamics based upon a hidden dynamic speech model in the form of segmental switching state space model that employs model parameters including those describing the unobserved speech dynamics and those describing the relationship between the unobserved speech dynamic vector and the observed acoustic feature vector is provided. The model parameters are modified based, at least in part, upon, a variational learning technique. In accordance with an aspect of the present invention, novel and powerful variational expectation maximization (EM) algorithm(s) for the segmental switching state space models used in speech applications, which are capable of capturing key internal (or hidden) dynamics of natural speech production, are provided. For example, modification of model parameters can be based upon an approximate mixture of Gaussian (MOG) posterior and/or based upon an approximate hidden Markov model (HMM) posterior using a variational technique.
REFERENCES:
patent: 5440661 (1995-08-01), Papcun
patent: 5625748 (1997-04-01), McDonough et al.
patent: 6052662 (2000-04-01), Hogden
patent: 6249763 (2001-06-01), Minematsu
patent: 6539353 (2003-03-01), Jiang et al.
patent: 6622119 (2003-09-01), Ramaswamy et al.
patent: 6629073 (2003-09-01), Hon et al.
patent: 6633857 (2003-10-01), Tipping
Ma et al, “A MIxture Linear Model with Target-Directed Dynamics for Spontaneous Speech Recognition”, IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP'02, vol. 1, pp. I-961-I-964.
Richards et al, “Vocal Tract Shape Trajectory Estimation Using MLP Analysis-by-Synthesis”, IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP-97, vol. 2, pp. 1287-1290.
Picone et al, “Initial Evaluation of Hidden Dynamic Models on Conversational Speech”, IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP'99, vol. 1, pp. 109-112.
M. Ostendorf, V. V. Digalakis, and O. A. Kimball, From HMMs to Segment Models: A Unified View of Stochastic Modeling for Speech Recognition, IEEE Trans. Speech Audio Process, vol. 4, No. 5, 1996. pp. 360-378.
Z. Ghahramani and G. E. Hinton, Variational Learning for Switching State-Space Models, Submitted to Neural Computation, vol. 12, 2000. pp. 831-864.
H. B. Richards and J. S. Bridle, The HDM: A Segmental Hidden Dynamic Model of Coarticulation, Proc. ICASSP, Phoenix, 1999, pp. 357-360.
K. Reinhard and M. Niranjan, Diphone subspace mixture trajectory models for HMM complementation, Speech Communications, vol. 38, 2002, pp. 237-265.
V. Pavlovic, B. J. Frey and T. S. Huang, Variational Learning in Mixed-State Dynamic Graphical Models, in Proc. UAI, Stockholm, 1999, pp. 522-530.
L. J. Lee, P. Fieguth and L. Deng, A Functional Articulatory Dynamic Model for Speech Production, in Proc. ICASSP, Salt Lake City, 2001, pp. 797-800.
J. Z. Ma and L. Deng, A path stacked algorithm for optimizing dynamic regimes in a statistical hidden dynamic model of speech, Computer, Speech and Languages, vol. 14, 2000, pp. 101-114.
L. Deng and J. Ma, Spontaneous speech recognition using a statistical coarticulatory model for the vocal-tract-resonance dynamics, J. Acous. Soc. Am., vol. 108, No. 6, 2000, pp. 3036-3048.
Attias Hagai
Deng Li
Lee Leo J.
Amin Turocy & Calvin LLP
Dorvil Richemond
Microsoft Corporation
Siedler Dorothy S
LandOfFree
Variational inference and learning for segmental switching... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Variational inference and learning for segmental switching..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Variational inference and learning for segmental switching... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-4042566