Method and system for pitch contour quantization in audio coding

• US 8,380,496 B2
• Filed: 04/25/2008
• Issued: 02/19/2013
• Est. Priority Date: 10/23/2003
• Status: Active Grant

• Current Assignee: RPX Corporation
• Original Assignee: Nokia Corporation
• Inventors: Nurminen, Jani, Himanen, Sakari, Heikkinen, Ari, Rm, Anssi
• Primary Examiner(s): He, Jialong
• Application Number:
  US12/150,307
• Publication Number:
  US 20080275695A1
• Time in Patent Office: 1,761 Days
• Field of Search: 704206-207, 704/220, 704/222, 704/219, 704/223, 704/500, 704/501
• US Class Current: 704/207
• CPC Class Codes: G10L 19/032 Quantisation or dequantisation of spectral components
• CPC Class Codes: G10L 19/09 Long term prediction, i.e. removing periodical redundancies, e.g. by using adaptive codebook or pitch predictor

Abstract

A method and device for improving coding efficiency in audio coding. From the pitch values of a pitch contour of an audio signal, a plurality of simplified pitch contour segments are generated to approximate the pitch contour, based on one or more pre-selected criteria. The contour segments can be linear or non-linear with each contour segment represented by a first end point and a second end point. If the contour segments are linear, then only the information regarding the end points, instead of the pitch values, are provided to a decoder for reconstructing the audio signal. The contour segment can have a fixed maximum length or a variable length, but the deviation between a contour segment and the pitch values in that segment is limited by a maximum value.

Claims

1. A method for coding an audio signal, comprising:

receiving pitch contour data indicative of the audio signal, the pitch contour data comprising a plurality of pitch values obtained from an audio segment at a plurality of sampling points at regular time intervals;

creating, in response to the pitch contour data obtained at said regular time intervals, a plurality of pitch contour segment candidates, each segment candidate corresponding to a sub-segment of the audio signal, wherein each sub-segment has a start-point pitch value and an end-point pitch value selected from said plurality of pitch values and each segment candidate has a start-segment pitch value at a start segment point and an end-segment pitch value at an end segment point, the start segment point aligned with the sampling point of the start-point pitch value and the end segment point aligned with the sampling point of the end-point pitch value;

measuring deviation between each of the pitch contour segment candidates and said pitch values in the corresponding sub-segment;

selecting, among said segment candidates, a plurality of consecutive simplified contour segments to represent the audio segment based on the measured deviations and one or more pre-selected criteria, wherein the start-segment pitch values at the start segment points of at least some simplified contour segments are different from the start-point pitch values of the corresponding sub-segments and the end-segment pitch values at the end segment points of at least some simplified contour segments are different from the end-point pitch values of the corresponding sub-segments, wherein each of the simplified contour segments is selected from a corresponding group of segment candidates, and wherein the simplified contour segments comprise a first contour segment and a plurality of subsequent contour segments, and wherein, in said creating,

the start-segment pitch value of the group of segment candidates corresponding to each of the subsequent contour segments is the same as the end-segment pitch value of the simplified contour segment immediately preceding said each of the subsequent contour segments, and

the start-segment pitch value of the group of segment candidates corresponding to the first contour segment is selected based on the start-segment pitch value of the sub-segment corresponding to first contour segment, and wherein the sub-segment corresponding to the first contour segment is representative of the pitch contour data first available after an inactive or unvoiced speech or at the beginning of an encoding process;

coding the sub-segment of the audio signal corresponding to the simplified contour segment with characteristics of the simplified contour segment.

2. The method according to claim 1, wherein each of the start segment points in said plurality of consecutive simplified contour segments is separated by a time distance from an immediately following start segment point, and wherein said coding comprises providing information indicative of the start-segment pitch value at the start segment point and the time distance between the start segment point and the immediately following start segment point so as to allow a decoder to reconstruct the audio signal in the audio sub-segment based on the information.

3. The method according to claim 2, wherein at least one of the selected candidates is a linear segment.

4. The method according to claim 2, wherein at least one of the selected candidates is a non-linear segment.

5. The method according to claim 1, wherein the number of pitch values in some of the consecutive sub-segments is equal to or greater than 3.

6. The method according to claim 1, wherein said creating is limited by a pre-selected condition such that the deviation between each of the simplified pitch contour segment candidates and each of said pitch values in the corresponding sub-segment is smaller than or equal to a pre-determined maximum value.

7. The method according to claim 6, wherein the created segment candidates have various lengths, and said selecting is based on the lengths of the segment candidates, and the pre-selected criteria include that

the selected candidate has the maximum length among the segment candidates.

8. The method according to claim 6, wherein said selecting is based on the lengths of the segment candidates, and the pre-selected criteria include that

the measured deviation is minimum among a group of the candidates having the same length.

9. The method according to claim 1, wherein said creating is carried out by adjusting the end segment point of the segment candidates.

10. The method according to claim 1, wherein the audio signal comprises a speech signal.

16. A non-transitory computer readable storage medium embodied with a software program for use in an encoding module, said software program comprising programming codes, when executed by a processor, perform the method according to claim 1.

11. An apparatus comprising:

an input end for receiving pitch contour data, the pitch contour data comprising a plurality of pitch values obtained from an audio segment of an audio signal at a plurality of sampling points at regular time intervals; and

a data processing module, responsive to the pitch contour data obtained from said regular time intervals, for generating a plurality of pitch contour segment candidates, each segment candidate corresponding to a sub-segment of the audio signal, wherein each sub-segment has a start-point pitch value and an end-point pitch value selected from said plurality of pitch values and each segment candidate has a start-segment pitch value at a start segment point and an end-segment pitch value at an end segment point, the start segment point aligned with the sampling point of the start-point pitch value and the end segment point aligned with the sampling point of the end-point pitch value, and wherein the processing module is configured to measure deviation between each of the pitch contour segment candidates and said pitch values in the corresponding sub-segment; and

to select, among said segment candidates, a plurality of consecutive simplified contour segments to represent the audio segment based on the measured deviations and pre-selected criteria, wherein the start-segment pitch values at the start segment points of at least some selected segment candidates are different from the start-point pitch values of the corresponding sub-segments and the end-segment pitch values at the end segment points of at least some simplified contour segments are different from the end-point pitch values of the corresponding sub-segments, wherein each of the simplified contour segments is selected from a corresponding group of segment candidates, and wherein the simplified contour segments comprise a first contour segment and a plurality of subsequent contour segments, and wherein, in said generating,

the start-segment pitch value of the group of segment candidates corresponding to each of the subsequent contour segments is the same as the end-segment pitch value of the simplified contour segment immediately preceding said each of the subsequent contour segments, and

the start-segment pitch value of the group of segment candidates corresponding to the first contour segment is selected based on the start-segment pitch value of the sub-segment corresponding to first contour segment, and wherein the sub-segment corresponding to the first contour segment is representative of the pitch contour data first available after an inactive or unvoiced speech or at the beginning of an encoding process.

12. The apparatus according to claim 11, further comprising

a quantization module configured to code the sub-segment of the audio signal corresponding to the simplified contour segment with characteristics of the simplified contour segment.

13. The apparatus according to claim 12, wherein the quantization module also configured to provide audio data indicative of the coded sub-segment, said coding device further comprising

a storage device, operatively connected to the quantization module to receive the audio data, for storing the audio data in a storage medium.

14. The apparatus according to claim 12, further comprising an output end, operatively connected to a storage medium, for providing audio data indicative of the coded sub-segment to the storage medium for storage.

15. The apparatus according to claim 12, further comprising an output end for transmitting audio data indicative of the coded sub-segment to the decoder so as to allow the decoder to reconstruct the audio signal based on the audio data.

17. An apparatus comprising:

an input for receiving audio data indicative of a plurality of consecutive simplified contour segments, the consecutive simplified contour segments selected from a plurality of pitch contour segment candidates, wherein the pitch contour segment candidates are generated in response to pitch contour data comprising a plurality of pitch values obtained from an audio segment of an audio signal at a plurality of sampling points at regular time intervals, each segment candidate corresponding to a sub-segment of the audio signal, wherein each sub-segment has a start-point pitch value and an end-point pitch value selected from said plurality of pitch values and each segment candidate has a start segment point and an end segment point, the start segment point aligned with the sampling point of the start-point pitch value and the end segment point aligned with the sampling point of the end-point pitch value, and wherein the plurality of consecutive simplified contour segments are selected among said segment candidates based on pre-selected criteria and on deviation between each of the segment candidate and said pitch values in the corresponding sub-segment, and wherein each of the simplified segments is defined by a first end point having a first pitch value and a second end point having a second pitch value, and wherein the first pitch values at the first end points of at least some simplified segments are different from the start-point pitch values of the corresponding sub-segments and the second pitch values at the second end points of at least some simplified segments are different from the end-point pitch values of the corresponding sub-segments, and wherein the received audio data comprises the end points defining the sub segments, wherein each of the simplified contour segments is selected from a corresponding group of segment candidates, and wherein the simplified contour segments comprise a first contour segment and a plurality of subsequent contour segments, and wherein, in generating the pitch contour segment candidates,

the start-segment pitch value of the group of segment candidates corresponding to each of the subsequent contour segments is the same as the end-segment pitch value of the simplified contour segment immediately preceding said each of the subsequent contour segments, and

the start-segment pitch value of the group of segment candidates corresponding to the first contour segment is selected based on the start-segment pitch value of the sub-segment corresponding to first contour segment, and wherein the sub-segment corresponding to the first contour segment is representative of the pitch contour data first available after an inactive or unvoiced speech or at the beginning of an encoding process; and

a reconstructing module configured to reconstruct the audio segment based on the received audio data.

18. The apparatus according to claim 17, wherein the audio data is recorded on an electronic media, and wherein the input is operatively connected to electronic media for receiving the audio data.

19. The apparatus according to claim 17, wherein the audio data is transmitted through a communication channel, and wherein the input is operatively connected to the communication channel for receiving the audio data.

20. The apparatus according to claim 17, comprising a mobile terminal.

21. A communication network, comprising:

a plurality of base stations; and

a plurality of mobile stations communicating with the base stations, wherein at least one of the mobile stations comprises;

an input for receiving audio data indicative of a plurality of consecutive simplified contour segments, the consecutive simplified contour segments selected from a plurality of pitch contour segment candidates, wherein the pitch contour segment candidates are generated in response to pitch contour data comprising a plurality of pitch values obtained from an audio segment of an audio signal at a plurality of sampling points at regular time intervals, each segment candidate corresponding to a sub-segment of the audio signal, wherein each sub-segment has a start-point pitch value and an end-point pitch value selected from said plurality of pitch values and each segment candidate has a start segment point and an end segment point, the start segment point aligned with the sampling point of the start-point pitch value and the end segment point aligned with the sampling point of the end-point pitch value, and wherein the plurality of consecutive simplified contour segments are selected among said segment candidates based on pre-selected criteria and on deviation between each of the segment candidate and said pitch values in the corresponding sub-segment, and wherein each of the simplified segments is defined by a first end point having a first pitch value and a second end point having a second pitch value, and wherein the first pitch values of the first end points of at least some simplified segments are different from the start-point pitch values of the corresponding sub-segments and the second pitch values of the second end points of at least some simplified segments are different from the end-point pitch values of the corresponding sub-segments, and wherein the received audio data comprises the end points defining the sub-segments, wherein each of the simplified contour segments is selected from a corresponding group of segment candidates, and wherein the simplified contour segments comprise a first contour segment and a plurality of subsequent contour segments, and wherein, in generating the pitch contour segment candidates,

the start-segment pitch value of the group of segment candidates corresponding to each of the subsequent contour segments is the same as the end-segment itch value of the simplified contour segment immediately preceding said each of the subsequent contour segments, and

the start-segment pitch value of the group of segment candidates corresponding to the first contour segment is selected based on the start-segment pitch value of the sub-segment corresponding to first contour segment, and wherein the sub-segment corresponding to the first contour segment is representative of the pitch contour data first available after an inactive or unvoiced speech or at the beginning of an encoding process; and

a reconstructing module configured to reconstruct the audio segment based on the received audio data.

22. An apparatus comprising:

means for receiving pitch contour data, the pitch contour data comprising a plurality of pitch values obtained from an audio segment of an audio signal at a plurality of sampling points at regular time intervals;

means, responsive to the pitch contour data obtained from said regular time intervals, for generating a plurality of pitch contour segment candidates, each segment candidate corresponding to a sub-segment of the audio signal, wherein each sub-segment has a start-point pitch value and an end-point pitch value selected from said plurality of pitch values and each segment candidate has a start-segment pitch value at a start segment point and an end-segment pitch value at an end segment point, the start segment point aligned with the sampling point of the start-point pitch value and the end segment point aligned with the sampling point of the end-point pitch value,

means, for measuring deviation between each of the pitch contour segment candidates and said pitch values in the corresponding sub-segment, and

means for selecting, among said segment candidates, a plurality of consecutive simplified contour segments to represent the audio segment based on the measured deviations and pre-selected criteria, wherein the start-segment pitch values of the start segment points of at least some selected segment candidates are different from the start-point pitch values of the corresponding sub-segments and the end-segment pitch values of the end segment points of at least some simplified contour segments are different from the end-point pitch values of the corresponding sub-segments, wherein each of the simplified contour segments is selected from a corresponding group of segment candidates, and wherein the simplified contour segments comprise a first contour segment and a plurality of subsequent contour segments, and wherein, in generating the plurality of pitch contour segment candidates,

the start-segment pitch value of the group of segment candidates corresponding to each of the subsequent contour segments is the same as the end-segment pitch value of the simplified contour segment immediately preceding said each of the subsequent contour segments, and

the start-segment pitch value of the group of segment candidates corresponding to the first contour segment is selected based on the start-segment pitch value of the sub-segment corresponding to first contour segment, and wherein the sub-segment corresponding to the first contour segment is representative of the pitch contour data first available after an inactive or unvoiced speech of at the beginning of an encoding process.

23. The apparatus according to claim 22, further comprising

means, responsive to the simplified contour segment, for coding the sub-segment of the audio signal corresponding to the simplified contour segment with characteristics of the selected simplified segment.