ENCODING APPARATUS, DECODING APPARATUS, AND METHODS

• US 20160293178A1
• Filed: 05/31/2016
• Published: 10/06/2016
• Est. Priority Date: 09/09/2011
• Status: Active Grant

• Current Assignee: Panasonic Intellectual Property Corporation of America (Panasonic Holdings Corporation)
• Original Assignee: Panasonic Intellectual Property Corporation of America (Panasonic Holdings Corporation)
• Inventors: DAIMOU, Katsunori, KAWASHIMA, Takuya, OSHIKIRI, Masahiro
• Granted Patent: US 9,741,356 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: G10L 19/265 Pre-filtering, e.g. high frequency emphasis prior to encoding
• CPC Class Codes: G10L 21/0388 Details of processing therefor

Abstract

A coding apparatus normalizes a low-frequency spectrum included in each of sub-bands obtained from dividing a low band part, using a largest amplitude value among the low-frequency spectrum included in each sub-band, obtains a normalized low-frequency spectrum by decoding the first encoded data, and calculates a correlation between each divided band of a high-frequency spectrum and a plurality of candidate bands of the normalized low-frequency spectrum. The best bands of a plurality of candidate bands are identified, each candidate band having a starting frequency position with non-zero amplitude in the normalized low-frequency spectrum, the high-frequency spectrum being in a high band part of the input audio signal that is higher than the predetermined frequency, and the high-frequency spectrum is encoded using lag information identifying the best band for transmitting the lag information to a decoder.

Claims

1. A coding apparatus, comprising:

a processor and a memory that stores instructions which, when executed by the processor, cause the processor to perform operations comprising;

encoding a low band part of an input audio signal to generate first encoded data which is transmitted to a decoder side, the low band part having a frequency equal to or lower than a predetermined frequency; and

normalizing a low-frequency spectrum included in each of sub-bands obtained by dividing the low band part, using a largest amplitude value among the low-frequency spectrum included in each sub-band,

obtaining a normalized low-frequency spectrum, the low-frequency spectrum being obtained by decoding the first encoded data;

calculating a correlation between each divided band of a high-frequency spectrum and a plurality of candidate bands of the normalized low-frequency spectrum;

identifying best bands of the plurality of candidate bands, each candidate band having a starting frequency position with non-zero amplitude in the normalized low-frequency spectrum, the high-frequency spectrum being a spectrum in a high band part of the input audio signal that is higher than the predetermined frequency; and

encoding the high-frequency spectrum using lag information identifying the best band for transmitting the lag information to the decoder side.

2. The coding apparatus according to claim 1,

wherein a plurality of other bands which have starting frequency positions with zero amplitude in the normalized low-frequency spectrum are skipped in calculating the correlation.

3. The coding apparatus according to claim 1,

wherein the lag information indicates a starting frequency position of the best band.

4. The coding apparatus according to claim 1,

wherein each band width of the plurality of candidate bands has a same width or a narrower width than each divided band of a high-frequency spectrum.

5. The coding apparatus according to claim 1,

wherein a number of the plurality of candidate bands is four.

6. A decoding apparatus, comprising:

a processor and a memory that stores instructions which, when executed by the processor, cause the processed to perform operations comprising;

decoding first encoded data, generated by encoding a low band part of an input audio signal, to generate a low-frequency spectrum, the low band part having a frequency equal to or lower than a predetermined frequency;

normalizing a low-frequency spectrum included in each of sub-bands obtained by dividing the low band part, using a largest amplitude value among the low-frequency spectrum included in each sub-band,

obtaining a normalized low-frequency spectrum, the low-frequency spectrum being obtained by decoding the first encoded data;

obtaining lag information identifying the best band; and

generating a high-frequency spectrum by using the best band;

wherein

the lag information indicating the particular band having a best correlation between each divided band of an encoding-side high-frequency spectrum and an encoding-side low-frequency spectrum, that results from normalizing a spectrum generated by decoding the first encoded data, the best bands being selected from a plurality of candidate bands, each having a starting frequency position with non-zero amplitude in the normalized low-frequency spectrum, the high-frequency spectrum is a high band part of the input audio signal that is higher than the predetermined frequency in the coding apparatus.

7. The decoding apparatus according to claim 6,

wherein the lag information indicates the starting frequency position of the best band.

8. The decoding apparatus according to claim 6,

wherein the second decoder identifies the particular band based on the lag information, and copies the best band to the high-frequency spectrum.

9. A coding method, comprising:

encoding, a low band part of an input audio signal to generate first encoded data which is transmitted to a decoder side, the low band part having a frequency equal to or lower than a predetermined frequency;

normalizing a low-frequency spectrum included in each of sub-bands obtained by dividing the low band part, using a largest amplitude value among the low-frequency spectrum included in each sub-band,

obtaining a normalized low-frequency spectrum, the low-frequency spectrum being obtained by decoding the first encoded data;

calculating a correlation between each divided band of a high-frequency spectrum and a plurality of candidate bands of the normalized low-frequency spectrum;

identifying best bands of the plurality of candidate bands, each candidate band having a starting frequency position with non-zero amplitude in the normalized low-frequency spectrum, the high-frequency spectrum being a spectrum in a high band part of the input audio signal that is higher than the predetermined frequency; and

encoding the high-frequency spectrum using lag information identifying the best band for transmitting the lag information to the decoder side.

10. The coding method according to claim 9,

wherein in calculating the correlation, a plurality of other bands which have starting frequency positions with zero amplitude in the normalized low-frequency spectrum are skipped.

11. The coding method according to claim 9,

wherein the lag information indicates a starting frequency position of the best band.

12. The coding method according to claim 9,

wherein each band width of the plurality of candidate bands has a same width or a narrower width than each divided band of a high-frequency spectrum.

13. The coding method according to claim 9,

wherein a number of the plurality of candidate bands is four.

14. A decoding method, comprising:

decoding first encoded data generated by encoding a low band part of an input audio signal to generate a low-frequency spectrum, the low band part having a frequency equal to or lower than a predetermined frequency; and

normalizing a low-frequency spectrum included in each of sub-bands obtained by dividing the low band part, using a largest amplitude value among the low-frequency spectrum included in each sub-band, and

obtaining a normalized low-frequency spectrum, the low-frequency spectrum being obtained by decoding the first encoded data;

decoding a second encoded data, and obtaining lag information identifying the best band; and

generating a high-frequency spectrum by using the best band;

wherein

the second encoded data contains lag information indicating the particular band where a best correlation between each divided band of an encoding-side high-frequency spectrum and an encoding-side low-frequency spectrum, that results from normalizing a spectrum generated by decoding the first encoded data the best bands being selected from a plurality of candidate bands, each having a starting frequency position with non-zero amplitude in the normalized low-frequency spectrum, the high-frequency spectrum is a high band part of the input audio signal that is higher than the predetermined frequency in the coding apparatus.

15. The decoding method according to claim 14,

wherein the lag information indicates the starting frequency position of the best band.

16. The decoding method according to claim 14, further comprising

identifying, using the second decoder, the particular band based on the lag information, and copying the best band to the high-frequency spectrum.