Discrete-Mixture HMMs-based Approach for Noisy Speech Recognition
Tetsuo Kosaka, Masaharu Katoh, Masaki Koh · 2007
This chapter introduced a new method of robust speech recognition using discrete-mixture HMMs (DMHMMs) based on maximum a posteriori (MAP) estimation. The aim of this work was to develop robust speech recognition for adverse conditions which contain both stationary and non-stationary noise. In order to achieve the goal, we proposed two methods.