fast codebook generation algorithm for speech signal

Hi, the following is the attachment of my code of fcg and also algorithm . can anyone plz help in writting code according to following algorithm . thanks in advance

Kekre’s Fast Codebook Generation Algorithm (KFCG) 1. Initially we have only one cluster which is the entire set of training vectors. Design a 1-vector codebook; which is the centroid of the cluster. 2. Split the cluster into two by comparing the first element of all the training vectors in the cluster with the first element of the centroid as follows: 3. If vi, 1>c1, 1 then vi,1 is grouped into C1 (cluster 1). 4. Else vi,1 is grouped into C2 (cluster 2). 5. Where v is the training vector and c is the centroid. 6. Find the centroids of C1 and C2 (this is 2-vector codebook). Now split C1 into two clusters by comparing the second element of all the training vectors in C1 with the second element of its centroids explained in step 2 above. Similarly split C2 into two clusters by comparing the second element of all the training vectors in C2 with the second element of its centroid. 7. Now four clusters are formed. Centroids of these four clusters are computed (this is 4-vector codebook). These four clusters are split further by comparing the third element of the training vectors in that cluster with the third element of its centroid as explained in step 2 above. 8. The process is repeated until a codebook of size M is designed.