Fast Lexically Constrained Viterbi Algorithm (FLCVA): Simultaneous Optimization of Speed and Memory

Lexical constraints on the input of speech and on-line handwriting systems improve the performance of such systems. A significant gain in speed can be achieved by integrating in a digraph structure the different Hidden Markov Models (HMM) corresponding to the words of the relevant lexicon. This integration avoids redundant computations by sharing intermediate results between HMM's corresponding to different words of the lexicon. In this paper, we introduce a token passing method to perform simultaneously the computation of the a posteriori probabilities of all the words of the lexicon. The coding scheme that we introduce for the tokens is optimal in the information theory sense. The tokens use the minimum possible number of bits. Overall, we optimize simultaneously the execution speed and the memory requirement of the recognition systems.

[1]  Frédéric Maire,et al.  Reduction of Non Deterministic Automata for Hidden Markov Model Based Pattern Recognition Applications , 2003, Australian Conference on Artificial Intelligence.

[2]  Edward Fredkin,et al.  Trie memory , 1960, Commun. ACM.

[3]  Steve Young,et al.  Token passing: a simple conceptual model for connected speech recognition systems , 1989 .

[4]  Nikos Fakotakis,et al.  Two Algorithms for Incremental Construction of Directed Acyclic Word Graphs , 1995, Int. J. Artif. Intell. Tools.

[5]  Dominique Revuz.  Dictionnaires et lexiques. Méthodes et algorithmes , 1991 .

[6]  Frederic Maire,et al.  A FAST LEXICALLY CONSTRAINED VITERBI ALGORITHM FOR ON­ LINE HANDWRITING RECOGNITIO , 2004 .

[7]  Robert Sedgewick,et al.  Algorithms in C++ - part 5: graph algorithms (3. ed.) , 2014 .

[8]  Biing-Hwang Juang,et al.  Fundamentals of speech recognition , 1993, Prentice Hall signal processing series.

[9]  Mehryar Mohri,et al.  An efficient algorithm for the n-best-strings problem , 2002, INTERSPEECH.

[10]  Robert Sedgewick,et al.  Algorithms in C : Part 5 : Graph Algo-rithms , 2002 .

[11]  Andrew J. Viterbi,et al.  Error bounds for convolutional codes and an asymptotically optimum decoding algorithm , 1967, IEEE Trans. Inf. Theory.

[12]  Sean R Eddy,et al.  What is dynamic programming? , 2004, Nature Biotechnology.

[13]  Renato De Mori,et al.  Lexical tree compression , 1991, EUROSPEECH.

[14]  1.1 introduction. , 2007, European journal of gastroenterology & hepatology.