In this paper an on-line handwriting recognition system with focus on adaptation techniques is described. Our Hidden Markov Model (HMM)-based recognition system for cursive German script can be adapted to the writing style of a new writer using either a retraining depending on the maximum likelihood (ML)-approach or an adaptation according to the maximum a posteriori (MAP)-criterion. The performance of the resulting writer-dependent system increases significantly, even if only a few words are available for adaptation. So this approach is also applicable for on-line systems in hand-held computers such as PDAs. This paper deals with the performance comparison of two different adaptation techniques either in a supervised or an unsupervised mode with the availability of different amounts of adaptation data ranging from only 6 words up to 100 words per writer.

Keywords: online cursive handwriting recognition, writer independent, adaptation of HMMs