Rational Interpolation of Maximum Likelihood Predictors Instochastic Language
نویسندگان
چکیده
In our paper, we address the problem of estimating stochastic language models based on n-gram statistics. We present a novel approach, rational interpolation, for the combination of a competing set of conditional n-gram word probability predictors, which consistently outper-forms the traditional linear interpolation scheme. The superiority of rational interpolation is substantiated by experimental results from language modeling, speech recognition , dialog act classiication, and language identiica-tion.
منابع مشابه
An Optimal G^2-Hermite Interpolation by Rational Cubic Bézier Curves
In this paper, we study a geometric G^2 Hermite interpolation by planar rational cubic Bézier curves. Two data points, two tangent vectors and two signed curvatures interpolated per each rational segment. We give the necessary and the sufficient intrinsic geometric conditions for two C^2 parametric curves to be connected with G2 continuity. Locally, the free parameters w...
متن کاملDiscriminative estimation of interpolation parameters for language model classifiers
In this paper we present a new approach for estimating the interpolation parameters of language models (LM) which are used as classifiers. With the classical maximum likelihood (ML) estimation theoretically one needs to have a huge amount of data and the fundamental density assumption has to be correct. Usually one of these conditions is violated, so different optimization techniques like maxim...
متن کاملThe Utility of Semantic-Pragmatic Information and Dialogue-State for Speech Recognition in Spoken Dialogue Systems
Information about the dialogue-state can be integrated into language models to improve performance of the speech recognizer in a dialogue system. A dialogue state is deened in this paper as the question , the user is replying to. One of the main problems in dialogue-state dependent language modelling is the limitation of training data. In order to obtain robust models, we use the method of rati...
متن کاملFeature Displacement Interpolation
Given a sparse set of feature matches, we want to compute an interpolated dense displacement map. The application may be stereo disparity computation, ow computation, or non-rigid medical registration. Also estimation of missing image data, may be phrased in this framework. Since the features often are very sparse, the interpolation model becomes crucial. We show that a maximum likelihood estim...
متن کاملBayesian Language Model Interpolation for Mobile Speech Input
This paper explores various static interpolation methods for approximating a single dynamically-interpolated language model used for a variety of recognition tasks on the Google Android platform. The goal is to find the statically-interpolated firstpass LM that best reduces search errors in a two-pass system or that even allows eliminating the more complex dynamic second pass entirely. Static i...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1997