Title | ||
---|---|---|
Factor analysis based session variability compensation for Automatic Speech Recognition |
Abstract | ||
---|---|---|
In this paper we propose a new feature normalization based on Factor Analysis (FA) for the problem of acoustic variability in Automatic Speech Recognition (ASR). The FA paradigm was previously used in the field of ASR, in order to model the usefull information: the HMM state dependent acoustic information. In this paper, we propose to use the FA paradigm to model the useless information (speaker- or channel-variability) in order to remove it from acoustic data frames. The transformed training data frames are then used to train new HMM models using the standard training algorithm. The transformation is also applied to the test data before the decoding process. With this approach we obtain, on french broadcast news, an absolute WER reduction of 1.3%. |
Year | DOI | Venue |
---|---|---|
2011 | 10.1109/ASRU.2011.6163920 | Automatic Speech Recognition and Understanding |
Keywords | Field | DocType |
hidden Markov models,speech coding,speech recognition,ASR,FA paradigm,HMM state dependent acoustic information,WER reduction,acoustic data frames,acoustic variability,automatic speech recognition,decoding process,factor analysis,session variability compensation | Speech processing,Speech coding,Normalization (statistics),Pattern recognition,Voice activity detection,Computer science,Speech recognition,Speaker recognition,Test data,Artificial intelligence,Hidden Markov model,Acoustic model | Conference |
ISBN | Citations | PageRank |
978-1-4673-0366-8 | 2 | 0.37 |
References | Authors | |
9 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mickael Rouvier | 1 | 2 | 0.37 |
Mohamed Bouallegue | 2 | 38 | 6.13 |
Driss Matrouf | 3 | 3 | 1.08 |
Georges Linares | 4 | 87 | 19.73 |