Title :
Simplified Scoring Methods in HMM Based Speech Recognition
Author :
Paramonov, Pavel ; Sutula, Nadezhda
Author_Institution :
Inst. of Automatics & Comput. Eng., Nat. Res. Univ. Moscow Power Eng. Inst., Moscow, Russia
Abstract :
In this paper we present two ways to simplify scoring in Hidden Markov Models (HMM) based speech recognition in order to reduce it´s computational complexity. We tested proposed approaches on Russian words recognition and compared results with those that gave recognition using conventional forward algorithm.
Keywords :
computational complexity; hidden Markov models; speech recognition; HMM based speech recognition; Russian word recognition; computational complexity; hidden Markov model based speech recognition; scoring methods; Computational complexity; Computational modeling; Hardware; Heuristic algorithms; Hidden Markov models; Speech; Speech recognition; Hidden Markov Models; forward algorithm; isolated words recognition; speech recognition;
Conference_Titel :
Soft Computing and Machine Intelligence (ISCMI), 2014 International Conference on
DOI :
10.1109/ISCMI.2014.32
