10.1184/R1/6473474.v1 Yajie Miao Yajie Miao Florian Metze Florian Metze Alexander Waibel Alexander Waibel Learning discriminative basis coefficients for eigenspace MLLR unsupervised adaptation Carnegie Mellon University 2013 Speaker adaptation discriminative training speech recognition 2013-05-01 00:00:00 Journal contribution https://kilthub.cmu.edu/articles/journal_contribution/Learning_discriminative_basis_coefficients_for_eigenspace_MLLR_unsupervised_adaptation/6473474 <p>Eigenspace MLLR is effective for fast adaptation when the amount of adaptation data is limited, e.g., less than 5s. The general motivation is to represent the MLLR transform as a linear combination of basis matrices. In this paper, we present a framework to estimate a speaker-independent discriminative transform over the combination coefficients. This discriminative basis coefficients transform (DBCT) is learned by optimizing discriminative criteria over all the training speakers. During recognition, the ML basis coefficients for each testing speaker are firstly found, on which DBCT is applied to give the final MLLR transform discrimination ability. Experiments show that DBCT results in consistent WER reduction in unsupervised adaptation, compared with both standard ML and discriminatively trained transforms.</p>