Method for reduction of aliasing introduced by spectral envelope adjustment in real-valued filterbanks

• US 10,685,661 B2
• Filed: 08/07/2019
• Issued: 06/16/2020
• Est. Priority Date: 09/18/2002
• Status: Expired due to Term

• Current Assignee: Dolby International AB (Dolby Laboratories Incorporated)
• Original Assignee: Dolby International AB (Dolby Laboratories Incorporated)
• Inventors: Kjoerling, Kristofer, Villemoes, Lars
• Primary Examiner(s): Colucci, Michael
• Application Number:
  US16/534,685
• Publication Number:
  US 20190362729A1
• Time in Patent Office: 314 Days
• Field of Search: 704232, 704219, 704 25, 704203, 708404, 375350, 375316, 375296, 375260, 375259, 375244, 375240, 341123
• CPC Class Codes: G10L 19/00 Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis in musical instruments G10H
• CPC Class Codes: G10L 19/0204 using subband decomposition
• CPC Class Codes: G10L 19/025 Detection of transients or attacks for time/frequency resolution switching
• CPC Class Codes: G10L 19/083 the excitation function being an excitation gain G10L25/90 takes precedence
• CPC Class Codes: G10L 19/12 the excitation function being a code excitation, e.g. in code excited linear prediction [CELP] vocoders
• CPC Class Codes: G10L 19/26 Pre-filtering or post-filtering
• CPC Class Codes: G10L 25/12 the extracted parameters being prediction coefficients
• CPC Class Codes: G10L 25/18 the extracted parameters being spectral information of each sub-band
• CPC Class Codes: G10L 25/21 the extracted parameters being power information
• CPC Class Codes: G10L 25/48 specially adapted for particular use
• CPC Class Codes: H03H 17/0266 Filter banks

Abstract

The present invention proposes a new method for improving the performance of a real-valued filterbank based spectral envelope adjuster. By adaptively locking the gain values for adjacent channels dependent on the sign of the channels, as defined in the application, reduced aliasing is achieved. Furthermore, the grouping of the channels during gain-calculation, gives an improved energy estimate of the real valued subband signals in the filterbank.

Claims

1. An audio decoder for producing a full bandwidth audio signal having a lowband portion and a highband portion, the apparatus comprising:

a cosine modulated, real-valued analysis filterbank that receives a time domain decoded audio signal and produces a plurality of real-valued subband signals;

a high frequency reconstructor that regenerates at least some of the highband portion by copying one or more of the plurality of real-valued subband signals up to the highband portion;

an aliasing detector that identifies subband signals where aliasing created by spectral envelope adjustment of an audio signal may occur based at least in part on a linear predictor applied to at least some of the plurality of real-valued subband signals;

an energy estimator that estimates an energy of at least some of the plurality of copied real-valued subband signals;

an aliasing reducer that modifies a gain to be applied to at least some of the identified subbands signals based at least in part on the estimated energy; and

a real-valued synthesis filterbank that combines the plurality of real-valued subband signals with the high-band portion to produce the full bandwidth audio signal, the full bandwidth audio including real-valued time domain samples,

wherein the audio decoder is implemented at least in part with one or more hardware elements, the linear predictor is a second order linear predictor, and the linear predictor uses a covariance method of linear prediction.

2. A method performed in an audio decoder for producing a full bandwidth audio signal having a lowband portion and a highband portion, the apparatus comprising:

filtering a time domain decoded audio signal with a real-valued, cosine modulated analysis filterbank to produce a plurality of real-valued subband signals;

regenerating at least some of the highband portion by copying one or more of the plurality of real-valued subband signals up to the highband portion;

identifying subband signals where aliasing created by spectral envelope adjustment of an audio signal may occur based at least in part on a linear predictor applied to at least some of the plurality of real-valued subband signals;

estimating an energy of at least some of the plurality of copied real-valued subband signals;

modifying a gain to be applied to at least some of the identified subbands signals based at least in part on the estimated energy;

filtering the plurality of real-valued subband signals and the highband portion with a real-valued synthesis filterbank to produce the full bandwidth audio signal, the full bandwidth audio including real-valued time domain samples; and

outputting the full bandwidth audio signal,

wherein the method is implemented at least in part with one of more hardware elements, and

wherein the linear predictor is a second order linear predictor and the linear predictor uses a covariance method of linear prediction.