Influence of vector quantization on isolated word recognition
Vincent Fontaine, Henri Leich, Jean Hennebert, Michael Hasler · 1994
. Vector Quantization can be considered as a data compression technique. In the last few years, vector quantization has been increasingly applied to reduce problem complexity like pattern recognition. In speech recognition, discrete systems are developed to build up real-time systems. This paper presents original results by comparing the KMeans and the Kohonen approaches on the same recognition platform. Influence of some quantization parameters is also investigated. It can be observed through the results presented in this paper that the quantization quality has a significant influence on the recognition rates. Surprisingly, the Kohonen approach leads to better recognition results despite its poor distortion performance. 1. Introduction Vector Quantization is almost considered as a data compression technique by DSP researchers. As a matter of fact, a lot of work has been done in this field to improve the quality of the speech- and image coders. The last few years, vector quantization ...