Encoding method, encoder, program and recording medium

Patent metadata

Field	Value
Publication number	US-9552821-B2
Application number	US-201314394158-A
Country	US
Kind code	B2
Filing date	May 29, 2013
Priority date	May 30, 2012
Publication date	Jan 24, 2017
Grant date	Jan 24, 2017

Abstract

A value of gain is updated so that the greater the difference between the number of bits or estimated number of bits in a code obtained by encoding a string of integer value samples obtained by dividing each sample in a sample string derived from an input audio signal in a given interval by gain before the update and a predetermined number B of allocated bits, the greater the difference between the gain before the update and the updated gain. A gain code corresponding to the updated gain and an integer signal code obtained by encoding a string of integer value samples obtained by dividing each sample in the sample string by the gain are obtained.

First claim

What is claimed is:

1. An encoding method for an input sample string derived from an input audio signal in a given interval, the encoding method obtaining a gain code corresponding to a gain obtained by a gain update loop processing step of obtaining the gain by loop processing, and an integer signal code obtained by encoding a string of integer value samples obtained by dividing each sample in the sample string by the gain;
wherein the gain update loop processing step comprises:
a lower limit gain setting step of, when the number of bits or estimated number of bits in a code obtained by encoding a string of integer value samples obtained by dividing each sample in the sample string by a gain before an update is greater than a predetermined number B of allocated bits, setting the gain before the update as a lower limit g min of the gain;
an upper limit gain setting step of, when the number of bits or estimated number of bits in a code obtained by encoding a string of integer value samples obtained by dividing each sample in the sample string by the gain before the update is smaller than the predetermined number B of allocated bits, setting the gain before the update as an upper limit g max of the gain; and
a gain update step of setting a weighted mean of the upper limit of the gain and the lower limit of the gain as an updated gain and outputting the updated gain, where a weight based on at least the predetermined number B of allocated bits, the number c L of consumed-bits-at-lower-limit-setting and the number c U of consumed-bits-at-upper-limit-setting is assigned to at least one of the upper limit g max of the gain and the lower limit g min of the gain, wherein
the number c L of consumed-bits-with-lower-limit-setting is the number of the bits or the estimated number of the bits in the code obtained by encoding the string of the integer value samples obtained by dividing each sample in the sample string by the gain before the update when the number of the bits or the estimated number of the bits is greater than the predetermined number B; and
the number c U of consumed-bits-at-upper-limit-setting is the number of the bits or the estimated number of the bits in the code obtained by encoding the string of the integer value samples obtained by dividing each sample in the sample string by the gain before the update when the number of the bits or the estimated number of the bits is smaller than the predetermined number B.

2. An encoding method for a sample string derived from an input audio signal in a given interval, the encoding method comprising:
a quantization step of quantizing a value obtained by diving each sample in the sample string by a gain to obtain a quantized normalized sample string;
a variable-length encoding step of encoding the quantized normalized sample string by variable-length encoding to obtain a sample string code;
a gain expansion update step of setting a value greater than the gain as new gain;
a gain reduction update step of setting a value smaller than the gain as new gain; and
a determination step of, when the number of updates of the gain is equal to a predetermined number of updates, outputting the gain and the sample string code, when the number of updates of the gain is less than the predetermined number of updates and the number of consumed bits which is the number of bits in the sample string code is greater than a predetermined number of allocated bits, causing the gain expansion update step to be performed, and when the number of updates of the gain is less than the predetermined number of updates and the number of the consumed bits is smaller than the predetermined number of allocated bits, causing the gain reduction update step to be performed;
wherein the gain expansion update step comprises:
a lower limit gain setting step of, when the number of the consumed bits is greater than the predetermined number of allocated bits, setting a value of gain corresponding to the number of the consumed bits as a lower limit of gain; and
a gain expansion step of, when the number of the consumed bits is greater than the predetermined number of allocated bits and an upper limit of the gain has not been set, updating a value of the gain so that the greater a value of A-T, the greater amount by which the value of the gain before the update increases to a value of updated gain, and causing the quantization step to be performed, where the value of A-T represents the number A of some or all of the samples in the quantized normalized sample string minus the number T of quantized normalized samples corresponding to a truncated sample string code left after removing a truncation code corresponding to amount by which the number of the consumed bits exceeds the predetermined number of allocated bits from the sample string code; and
the gain reduction update step comprises:
an upper limit gain setting step of, when the number of the consumed bits is smaller than the predetermined number of allocated bits, setting a value of gain corresponding to the number of the consumed bits as an upper limit of gain; and
a gain reduction step of, when the number of the consumed bits is smaller than the predetermined number of allocated bits and a lower limit of the gain has not been set, updating the value of the gain so that the greater the predetermined number of allocated bits minus the number of the consumed bits, the greater amount by which the value of the gain before the update decreases to an updated value, and causing the quantization step to be performed.

3. An encoding method for a sample string derived from an input audio signal in a given interval, the encoding method comprising:
a quantization step of quantizing a value obtained by dividing each sample in the sample string by a gain to obtain a quantized normalized sample string;
a gain expansion update step of setting a value greater than the gain as new gain;
a gain reduction update step of setting a value smaller than the gain as new gain;
a variable-length encoding step of encoding the quantized normalized sample string by variable-length encoding to obtain a sample string code; and
a determination step of, when the number of updates of the gain is equal to a predetermined number of updates, causing the variable-length encoding step to be performed, when the number of updates of the gain is less than the predetermined number of updates and the number of consumed bits which is an estimated number of bits in a code corresponding to the quantized normalized sample string is greater than a predetermined number of allocated bits, causing the gain expansion update step to be performed, and when the number of updates of the gain is less than the predetermined number of updates and the number of the consumed bits is smaller than the predetermined number of allocated bits, causing the gain reduction update step to be performed;
wherein the gain expansion update step comprises:
a lower limit gain setting step of, when the number of the consumed bits is greater than the predetermined number of allocated bits, setting a value of gain corresponding to the number of the consumed bits as a lower limit of gain; and
a gain expansion step of, when the number of the consumed bits is greater than the predetermined number of allocated bits and an upper limit of the gain has not been set, updating a value of the gain so that the greater a value of A-T, the greater amount by which the value of the gain before the update increases to an updated value, and causing the quantization step to be performed, where the value of A-T represents the number A of some or all of the samples in the quantized normalized sample string minus the number T of samples left after removing quantized normalized samples from the quantized normalized sample string, the quantized normalized samples directed to truncation cod

Assignees

• Nippon Telegraph & Telephone

Inventors

• Moriya Takehiro
• Kamamoto Yutaka
• Harada Noboru
• Hiwasaki Yusuke
• Fukui Masahiro

Classifications

• G10L19/002

  Dynamic bit allocation (for perceptual audio coders G10L19/032) · CPC title

• G10L19/032 Primary

  Quantisation or dequantisation of spectral components · CPC title

• G10L19/083

  the excitation function being an excitation gain (G10L25/90 takes precedence) · CPC title

• G10L19/08

  Determination or coding of the excitation function; Determination or coding of the long-term prediction parameters · CPC title

• G10L19/035 Primary

  Scalar quantisation · CPC title