Language modeling using stochastic automata with variable length contexts Export

@article{citeulike:4006517, abstract = {It is well known that language models are effective for increasing the accuracy of speech and handwriting recognizers, but large language models are often required to achieve low model perplexity (or entropy) and still have adequate language coverage. We study three efficient methods for variable order stochastic language modeling in the context of the stochastic pattern recognition problem. Two of these methods are previous techniques from recent literature, and one is a new method based on a successful text compression technique. We give results of a comparative analysis, and demonstrate that the best performance is achieved by extending one of the previous techniques using elements from the newly developed method.}, author = {Hu, J.}, citeulike-article-id = {4006517}, citeulike-linkout-0 = {http://dx.doi.org/10.1006/csla.1996.0020}, citeulike-linkout-1 = {http://linkinghub.elsevier.com/retrieve/pii/S0885-2308(96)90020-0}, doi = {10.1006/csla.1996.0020}, issn = {08852308}, journal = {Computer Speech \& Language}, keywords = {language, length, model, nlp, variable}, month = {January}, number = {1}, pages = {1--16}, posted-at = {2009-02-04 11:33:44}, priority = {2}, title = {Language modeling using stochastic automata with variable length contexts}, url = {http://dx.doi.org/10.1006/csla.1996.0020}, volume = {11}, year = {1997} }

TY - JOUR ID - citeulike:4006517 L3 - citeulike-article-id:4006517 N2 - It is well known that language models are effective for increasing the accuracy of speech and handwriting recognizers, but large language models are often required to achieve low model perplexity (or entropy) and still have adequate language coverage. We study three efficient methods for variable order stochastic language modeling in the context of the stochastic pattern recognition problem. Two of these methods are previous techniques from recent literature, and one is a new method based on a successful text compression technique. We give results of a comparative analysis, and demonstrate that the best performance is achieved by extending one of the previous techniques using elements from the newly developed method. IS - 1 JF - Computer Speech & Language SN - 08852308 EP - 16 TI - Language modeling using stochastic automata with variable length contexts VL - 11 SP - 1 KW - language KW - length KW - model KW - nlp KW - variable AU - Hu, J PY - 1997/01// UR - http://dx.doi.org/10.1006/csla.1996.0020 ER -