5th International Conference on Spoken Language Processing

Sydney, Australia
November 30 - December 4, 1998

Using X-Gram For Efficient Speech Recognition

Antonio Bonafonte, Josť B. Marino

Universitat Politecnica de Catalunya, Spain

X-grams are a generalization of the n-grams, where the number of previous conditioning words is different for each case and decided from the training data. X-grams reduce perplexity with respect to trigrams and need less number of parameters. In this paper, the representation of the x-grams using finite state automata is considered. This representation leads to a new model, the non-deterministic x-grams, an approximation that is much more efficient, suffering small degradation on the modeling capability. Empirical experiments for a continuous speech recognition task show how, for each ending word, the number of transitions is reduced from 1222 (the size of the lexicon) to around 66.

Full Paper

Bibliographic reference.  Bonafonte, Antonio / Marino, Josť B. (1998): "Using x-gram for efficient speech recognition", In ICSLP-1998, paper 1125.