VOICE CODING METHOD

摘要

PROBLEM TO BE SOLVED: To improve compression ratio, in the case of predictive-coding of multi-channel voice signals, considering the fact that higher order bit data of the multi-channel voice signals each are closely correlated with each other while lower order bit data are not correlated. SOLUTION: An adding circuit 1a calculates a sum signal (L+R) of two- channel stereo signals L, R and outputs the signal to a lossless single channel encoder 2D1 for the sum channel, and a subtracting circuit 1b calculates the difference signal (L-R) and outputs it to a lossless single channel encoder 2D2 for the difference channel. The encoders 2D1, 2D2 perform predictive-coding of the differencesΔ(L+R) andΔ(L-R) of the sum signal (L+R) and the difference signal (L-R) of higher 16-bits among 24 bits, and also transmit lower order 8 bits as they are, via a recoding medium and a communication medium.