An Audio Object Codec Method Based on Spectrum Shifting

Abstract

The invention discloses an audio object coding and decoding method based on frequency spectrum shifting. This method proposes a strategy including global shifting and local shifting to reduce aliasing distortion. In the encoding stage, time-frequency transformation is first performed to obtain the spectrum information of multiple input signals; then, it is judged whether aliasing occurs in each subband, and the aliased areas will be sorted according to the degree of aliasing; after determining the time-frequency area that needs to be moved, The aliased time-frequency points are moved to the non-aliasing area, and there are two strategies for moving the time-divided overall shift and local shift. The overall shift can greatly reduce the shift information that needs to be recorded; finally, the downmix signal and side information are synthesized into a code stream. In the decoding stage, the time-frequency components are first restored to their original positions according to the moving information, and then decoded according to the joint audio object coding framework SAOC. The present invention moves aliased time-frequency information to a non-aliased area in a downmixing process by using a spectrum shifting strategy, thereby reducing aliasing distortion and improving decoded audio quality.

Description

technical field [0001] The invention belongs to the technical field of digital audio signal processing, and in particular relates to an audio object encoding and decoding method based on frequency spectrum shifting, which is applicable to simultaneously transmitted multi-audio object signals, and restores each audio object with good sound quality at the decoding end. Background technique [0002] Spatial audio technology, also known as 3D audio, has grown in popularity for its unparalleled immersive listening experience that cannot be matched by mono or stereo audio. Traditional spatial audio coding schemes are based on channels, but this method has certain limitations [Document 1] [Document 2]. First, it is difficult to be compatible with non-standard playback environments. When the number and placement of speakers does not meet standard placement requirements, the spatial perception of audio is distorted. Second, channel-based spatial audio is difficult to support intera...

AI Technical Summary

Problems solved by technology

Its improved method improves the listener's listening experience to a certain extent, but the listener can still perceive the problem of aliasing and distortion

Images