Minimum classification error training of hidden Markov models for handwriting recognition
نویسنده
چکیده
This paper evaluates the application of the Minimum Classification Error (MCE) training to online-handwritten text recognition based on Hidden Markov Models. We describe an allograph-based, character level MCE training aimed at minimizing the character error rate while enabling flexibility in writing style. Experiments on a writer-independent discrete character recognition task covering all alpha-numerical characters and keyboard symbols show that MCE achieves more than 30% character error rate reduction compared to the baseline Maximum Likelihood-based system.
منابع مشابه
Improved performance and generalization of minimum classification error training for continuous speech recognition
Discriminative training of hidden Markov models (HMMs) using segmental minimum classi cation error (MCE) training has been shown to work extremely well for certain speech recognition applications. It is, however, somewhat prone to overspecialization. This study investigates various techniques which improve performance and generalization of the MCE algorithm. Improvements of up to 7% in relative...
متن کاملSome problems with a Scotsman and a Russian: Caveats on Bayesian and hidden-Markov models (v2.3)
This paper describes a number of fundamental and practical problems in the application of hidden-Markov models and Bayes when applied to cursive-script recognition. Several problems, however, will have an effect in other application areas. The most fundamental problem is the propagation of error in the product of probabilities. This is a common and pervasive problem which deserves more attentio...
متن کاملMinimum Classification Error Training of Hidden Markov Models for Sequential Data in the Wavelet Domain
In the last years there has been increasing interest in developing discriminative training methods for hidden Markov models, with the aim to improve their performance in classification and pattern recognition tasks. Although several advances have been made in this area, they have been targeted almost exclusively to standard models whose conditional observations are given by a Gaussian mixture d...
متن کاملA log-linear discriminative modeling framework for speech recognition
Conventional speech recognition systems are based on Gaussian hidden Markov models (HMMs). Discriminative techniques such as log-linear modeling have been investigated in speech recognition only recently. This thesis establishes a log-linear modeling framework in the context of discriminative training criteria, with examples from continuous speech recognition, part-of-speech tagging, and handwr...
متن کاملImproved Modeling in Handwriting Recognition
In this work a script independent handwriting recognition system is proposed which is derived from the RWTH-ASR hidden Markov model (HMM) based speech recognizer. Most problems occurring in handwriting recognition (HWR) are induced by large variations within the written text. In particular, different handwriting styles such as cursive writing or long drawn-out strokes are difficult to model. Co...
متن کامل