DocumentCode :
312018
Title :
Using the self-organizing map to speed up the probability density estimation for speech recognition with mixture density HMMs
Author :
Kurimo, Mikko ; Somervuo, Panu
Author_Institution :
Neural Networks Res. Centre, Helsinki Univ. of Technol., Espoo, Finland
Volume :
1
fYear :
1996
fDate :
3-6 Oct 1996
Firstpage :
358
Abstract :
This paper presents methods to improve the probability density estimation in hidden Markov models for phoneme recognition by exploiting the self-organizing map (SOM) algorithm. The advantage of using the SOM is based on the created approximative topology between the mixture densities by training the Gaussian mean vectors used as the kernel centers by the SOM algorithm. The topology allows the neighboring mixtures to respond strongly to the same inputs and so most of the nearest mixtures used to approximate the current observation probability will be found in the topological neighborhood of the “winner” mixture. Also the knowledge about the previous winners are used to speed up the search for the new winners. Tree-search SOMs and segmental SOM training are studied aiming at faster search and suitability for HMM training. The framework for the presented experiments includes mel-cepstrum features and phoneme-wise tied mixture density HMMs
Keywords :
Gaussian processes; cepstral analysis; hidden Markov models; learning (artificial intelligence); probability; self-organising feature maps; speech recognition; tree searching; Gaussian mean vectors; SOM; mel-cepstrum features; mixture density hidden Markov model; neighboring mixtures; observation probability; phoneme recognition; phoneme-wise tied mixture density HMM; probability density estimation; searching; segmental SOM training; self-organizing map; speech recognition; training; tree search; Cepstrum; Hidden Markov models; Neural networks; Organizing; Samarium; Spatial databases; Speech recognition; Topology;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on
Conference_Location :
Philadelphia, PA
Print_ISBN :
0-7803-3555-4
Type :
conf
DOI :
10.1109/ICSLP.1996.607128
Filename :
607128
Link To Document :
بازگشت