| 摘要 |
PURPOSE:To attain voice coding of high quality by using the small volume of transmission information by processing a continuous sample sequence having an equal difference between adjacent samples as a block and compressing the voice sample sequence to code it. CONSTITUTION:The voice sample blocks 1 to 5 are continuous sequences having an equal difference between adjacent samples. The number of samples of the block 1 is '7', that of the blocks 2 to 4 is '2' and the block 5 is '4', the absolute values of differences in respective blocks are shown by '6'-'10' and each threshold is shown by '11'. The differences 6 to 9 of the blocks 1 to 3, 5 whose absolute value of the difference is less than the threshold '11' are coded. Since two or more samples are included in each block, the number of samples is also coded. In the block 4 whose absolute value of the difference is more than the threshold '11', the number '2' of samples is coded. Since the number of bits for difference coding is set up to a value less than the number of bits for sampling, the maximum difference to be displayed as a code is set up as a threshold and a sample value is coded in each sample when the absolute value of the difference exceeds the threshold, information can be efficiently compressed and voice coding of high quality can be attained by using the small volume of information transmission. |
