Novel Decimation-Whitening Filter in Spectral Band Replication
MPEG-4 high-efficiency advanced audio coding (HE-AAC) has adopted spectral band replication (SBR) to efficiently compress high-frequency parts of the audio. In SBR, the linear prediction is applied to low-frequency subbands to clip the tonal components and smooth the associated spectrum for replicating to high-frequency bands. Such a process is referred to as the whitening filtering. In SBR, to avoid the alias artifact from spectral adjustment, a complex filterbank instead of real filterbank is adopted. For the QMF subbands, this paper analyzes that the linear prediction defined in SBR standard results in the predictive biases. An new whitening filter, called the decimation-whitening filter, is proposed to eliminate the predictive bias and provide advantages in terms of noise-to-signal ratio measure, frequency resolution, energy leakage, and computational complexity for SBR.
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member and would like to subscribe to the E-Library then Join the AES!
This paper costs $33 for non-members and is temporarily free for AES members.