Robust Speech Recognitionfor Speaker and Speaking Environment

Provided by: International Journal of Engineering Trends and Technology
Topic: Software
Format: PDF
The authors study a cohort selection technique to construct a space with good resolution in the online phase. The concept of cohort selection resembles that of the family of subset selection methods that find a subset of components from the entire set of components to model a signal of interest. The ESSEM process comprises two phases, the offline and the online. In the offline phase, they prepare an ensemble speaker and speaking environment space formed by a collection of super-vectors. In the online phase, with the ensemble environment space prepared in the offline phase, they estimate the super-vector for a new testing environment based on a stochastic matching criterion.

Find By Topic