1: \begin{abstract}

2:   Hidden Markov model (HMM) has been successfully used for sequential

3:   data modeling problems.

4:   In this work, we propose to power the modeling capacity of HMM by

5:   bringing in neural network based generative models.

6:   The proposed model is termed as GenHMM.

7:   In the proposed GenHMM, each HMM hidden state is associated with a

8:   neural network based generative model that has

9:   tractability of exact likelihood and provides efficient likelihood

10:   computation.

11:   A generative model in GenHMM consists of a mixture of generators

12:   that are realized by flow models.

13:   A learning algorithm for GenHMM is proposed in

14:   expectation-maximization framework.

15:   The convergence of the learning GenHMM is analyzed.

16:   We demonstrate the efficiency of GenHMM by classification tasks on practical sequential data.

17: \end{abstract}

18: