Methods and apparatus for decreasing the size of pattern recognition

Abstract

Methods and apparatus for producing efficiently sized models suitable for pattern recognition purposes are described. Various embodiments are directed to the automated generation, evaluation, and selection of reduced size models from an initial model having a relatively large number of components, e.g., more components than can be stored for a particular intended application. To achieve model size reduction in an automated iterative manner, expectation maximization (EM) model training techniques are combined, in accordance with the present invention, with model size constraints. In one embodiment, a plurality of reduced size models are generated using a LaGrange multiplier from an input model and input size constraints. The plurality of reduced size models are stored in a buffer and scored using a likelihood scoring technique. The one of the reduced size models receiving the highest score may be selected as the reduced size model to be output or used as an input model during future iterations of the model size reduction process. The reduced size model to be used, e.g., for speech, image or other pattern recognition purposes, may be selected from the buffered models produced during multiple iterations of the model size reduction process.