Initialization of hidden Markov models for unconstrained on-line handwriting recognition

Authors:
K. Nathan;A. Senior;J. Subrahmonia
Affiliations:
IBM Thomas J. Watson Res. Center, Yorktown Heights, NY, USA;-;-
Venue:
ICASSP '96 Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE International Conference - Volume 06
Year:
1996

Citing 0
Cited 5

A Combination Fingerprint Classifier

IEEE Transactions on Pattern Analysis and Machine Intelligence - Graph Algorithms and Computer Vision
Combining Online and Offline Handwriting Recognition

ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 2
MS-TDNN with Global Discriminant Trainings

ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
Unsupervised parameter selection for gesture recognition with vector quantization and Hidden Markov models

INTERACT'11 Proceedings of the 13th IFIP TC 13 international conference on Human-computer interaction - Volume Part IV
Incremental word learning: Efficient HMM initialization and large margin discriminative adaptation

Speech Communication

Quantified Score

Hi-index	0.00

Visualization

Abstract

In a hidden Markov model system, the initialization of the model parameters is critical to the performance of the model after retraining. This paper proposes a number of new approaches to the problem of initialization, and demonstrates that a method of smooth alignment results in the best performance.