This thesis addresses both of these limitations. Moreno, P. CS Dept. However, these combination methods can be computationally expensive, as data is processed by multiple systems. This limits the allowed forms of diversities that the ensemble can have.
Liu, F. The proposed method maps the state cluster posteriors from the teachers' sets of state clusters to that of the student. When performing recognition, these ensembles can be combined at the hypothesis and frame levels.
The proposals in this thesis are assessed on four ASR tasks. Balakrishnan Abstract: The goal of the proposed study is robust speech feature prediction using mel-LPC to improve the performance of speech recognition in adverse conditions and compares the performance with those standard LPC and MFCC through English dictation system with 14, isolated words and 9, connected words.
Gerazov, B. However, many of these have computational costs that scale linearly with the ensemble size. Cirilus and Methodius, Skopje.
One method to address this is teacher-student learning, which compresses the ensemble into a single student.