The Codebook Design Based on Ant Colony Clustering Algorithm and its Application in AMR-WB

Codebook design plays a crucial role in the performance of speech signal processing systems based on vector quantization, such as speech codec, speech recognition and so on. LBG algorithm is one of the most effective methods to generate codebooks and is commonly used in multidimensional signal processing. But it has the problem of initialize codebook design. The paper proposed adopting ant colony clustering algorithm to design codebook. But the clustering number of ant colony clustering algorithm is automatically formed, it may not be exactly equal to the codebook size. So the paper put forward solving measures on the basis of nearest neighbor criterion or decomposition method. The codebook design procedures with ant colony clustering algorithm is given in the paper detailedly. Using the proposed method, the paper designed vector quantizer´ codebooks of adaptive multi-rate wideband (AMR-WB) speech codec. Simulation results showed that the decoding speech qualities of AMR-WB speech codec were not almost degraded adopting newly designed codebooks. The transparent quantization effect of vector quantization was obtained.