A new i-vector approach and its application to irrelevant variability normalization based acoustic model training

Yu Zhang, Zhi-Jie Yan, Qiang Huo
2011 2011 IEEE International Workshop on Machine Learning for Signal Processing  
This paper presents a new approach to extracting a lowdimensional i-vector from a speech segment to represent acoustic information irrelevant to phonetic classification. Compared with the traditional i-vector approach, a full factor analysis model with a residual term is used. New procedures for hyperparameter estimation and i-vector extraction are derived and presented. The proposed i-vector approach is applied to acoustic sniffing for irrelevant variability normalization based acoustic model
more » ... raining in large vocabulary continuous speech recognition. Its effectiveness is confirmed by experimental results on Switchboard-1 conversational telephone speech transcription task.
doi:10.1109/mlsp.2011.6064637 dblp:conf/mlsp/0007YH11 fatcat:2d6ceoss6vfjbldy65xbdeud3e