net.java.sip.communicator.impl.neomedia.codec.audio.silk
Class Silk_VQ_nearest_neighbor_FLP

java.lang.Object
  net.java.sip.communicator.impl.neomedia.codec.audio.silk.Silk_VQ_nearest_neighbor_FLP

public class Silk_VQ_nearest_neighbor_FLP
extends Object

Author:

Dingxin Xu

Constructor Summary

Silk_VQ_nearest_neighbor_FLP()

Method Summary

(package private) static void

SKP_Silk_VQ_WMat_EC_FLP(int[] ind, int ind_offset, float[] rate_dist, float[] in, int in_offset, float[] W, int W_offset, short[] cb, short[] cl_Q6, float mu, int L) entropy constrained MATRIX-weighted VQ, for a single input data vector.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

Silk_VQ_nearest_neighbor_FLP

public Silk_VQ_nearest_neighbor_FLP()

Method Detail

SKP_Silk_VQ_WMat_EC_FLP

static void SKP_Silk_VQ_WMat_EC_FLP(int[] ind,
                                    int ind_offset,
                                    float[] rate_dist,
                                    float[] in,
                                    int in_offset,
                                    float[] W,
                                    int W_offset,
                                    short[] cb,
                                    short[] cl_Q6,
                                    float mu,
                                    int L)

entropy constrained MATRIX-weighted VQ, for a single input data vector.

Parameters:

ind - Index of best codebook vector

rate_dist - Best weighted quant. error + mu * rate

in - Input vector to be quantized

W - Weighting matrix

cb - Codebook

cl_Q6 - Code length for each codebook vector

mu - Tradeoff between WSSE and rate

L - Number of vectors in codebook