Title :
Class-based Gaussian selection for efficient decoding in PTM HMMs
Author :
Son, J. ; Jung, S. ; Bae, K.
Author_Institution :
Sch. of Electron. & Electr. Eng., Kyungpook Nat. Univ., South Korea
Abstract :
A new Gaussian selection (GS) method is presented for fast decoding in phonetic tied-mixture (PTM) hidden Markov models (HMMs). For efficient likelihood computation, a constraint is imposed on the context-dependent weights as well as the number of Gaussians. Experimental results demonstrate the superiority of the proposed method over conventional GS methods.
Keywords :
Gaussian distribution; Gaussian processes; decoding; hidden Markov models; speech coding; speech recognition; GS methods; Gaussian constraint; PTM HMM; class-based Gaussian selection; context-dependent weights constraint; decoding; likelihood computation; phonetic tied-mixture hidden Markov models; speech recognition systems;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20040079
