Residual NOise Compensation by A Sequential EM Algorithm for robust Speech Recognition in Nonstationary Noise

摘要

We model noise as a stationary component plus a time varying residual. The stationary part is estiamted off-line and compensated using Log-Add noise compensation .The time varying residual is estiamted and compensated using a sequential EM algorithm. The residual noise compensation proceeds in parallel with the recognition process. Experimental results demonstrate that the proposed algorithm improves the recognition performance not only in highly non-stationary noise but also in slow-varying noise, compared with Log-Add noise compensation alone.