| 摘要 |
<p>An encoding method comprising: a normalization signal encoding process for obtaining normalization signal code obtained by encoding a sequence by means of a signal in which each input signal sample included in the input signal sequence has been normalized, and for obtaining a quantization normalized signal sequence corresponding to the normalization signal code; a global gain encoding process for obtaining a quantization global gain corresponding to the input signal sequence, and for obtaining a global gain code corresponding to the quantization global gain; a dividing process for dividing the quantization normalized signal sequence into a plurality of ranges; and a gain correction amount encoding process for obtaining a gain correction code which is for specifying a gain correction amount for each of the divided ranges and for which the correlation between the signal sequence and the input signal sequence is the highest, or the error is the lowest, said correlation being obtained by multiplying the corrected gain obtained by correcting the quantization global gain for each of the divided ranges by a gain correction amount, with the values for each sample of the quantization normalized signal sequence.</p> |
