Aggregate a posteriori linear regression adaptation

Jen-Tzung Chien*, Chih Hsien Huang

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

11 Scopus citations


We present a new discriminative linear regression adaptation algorithm for hidden Markov model (HMM) based speech recognition. The cluster-dependent regression matrices arc estimated from speaker-specific adaptation data through maximizing the aggregate a posteriori probability, which can be expressed in a form of classification error function adopting the logarithm of posterior distribution as the discriminant function. Accordingly, the aggregate a posteriori linear regression (AAPLR) is developed for discriminative adaptation where the classification errors of adaptation data arc minimized. Because the prior distribution of regression matrix is involved, AAPLR is geared with the Bayesian learning capability. We demonstrate that the difference between AAPLR discriminative adaptation and maximum a posteriori linear regression (MAPLR) adaptation is due to the treatment of the evidence. Different from minimum classification error linear regression (MCELR), AAPLR has closed-form solution to fulfill rapid adaptation. Experimental results reveal that AAPLR speaker adaptation docs improve speech recognition performance with moderate computational cost compared to maximum likelihood linear regression (MLLR), MAPLR, MCELR and conditional maximum likelihood linear regression (CMLLR). These results are verified for supervised adaptation as well as unsupervised adaptation for different numbers of adaptation data.

Original languageEnglish
Pages (from-to)797-807
Number of pages11
JournalIEEE Transactions on Audio, Speech and Language Processing
Issue number3
StatePublished - 1 May 2006


  • Aggregate a posteriori criterion
  • Bayesian learning
  • Discriminative adaptation
  • Linear regression adaptation
  • Speaker adaptation
  • Speech recognition

Fingerprint Dive into the research topics of 'Aggregate a posteriori linear regression adaptation'. Together they form a unique fingerprint.

Cite this