In this paper a method for analytic handwritten word recognition based on causal Markov random fields is described. The word models are hmms where each state corresponds to a letter; each letter is modeled by a nshp (Markov field). The word models are built dynamically. Letter and word model training is made using baum algorithm where the parameters are re-estimated on the generated word global models. The segmentation is not necessary~: the system determines itself during training the best repartition of the information within the letter models. First experiments on two real databases of french check amount words give very encouraging results up to 86% for recognition.