A Combined Rough Sets–K-means Vector Quantization Model for Arabic Speech Recognizer
Hanan Hassan Ali Adlan, A. B. D. RAHMAN RAMLI · 2016
Vector quantization (VQ), is considered an efficient data reduction technique, and is used as a preprocessing stage in speech recognition systems. Methods traditionally used for vector quantization are purely numerical methods rather than rule-based methods. Furthermore, most of the experiments performed in previous research work used English data sets, and fewer experiments on Arabic data. In this paper, a vector quantization model that incorporate rough sets attribute reduction and rules generation with a modified version of the K-means clustering algorithm was developed implemented and tested as a part of a speech recognition framework, the learning vector quantization.(LVQ) neural network model was used in the pattern matching stage. Arabic speech data was used in the original experiments, for both speaker dependant and speaker independent tests. The performance was found to be very high in terms of recognition time and