Audio Signal Processing and Coding

5.8 PERCEPTUAL BIT ALLOCATION

In this section, we will extend the uniform- and optimal-bit allocation algorithms presented in Chapter 3, Section 3.5, with perceptual bit-assignment strategies. In perceptual bit allocation method, the number of bits allocated to different bands is determined based on the global masking thresholds obtained from the psychoacoustic model. The steps involved in the computation of the global masking thresholds have been presented in detail in the previous section. The signal-to-mask ratio (SMR) determines the number of bits to be assigned in each band for perceptually transparent coding of the input audio. The noise-to-mask ratios (NMRs) are computed by subtracting the SMR from the SNR in each subband, i.e.,

The main objective in a perceptual bit allocation scheme is to keep the quantization noise below a masking threshold. For example, note that the NMR in Figure 5.11(a) is relatively more compared to NMR in Figure 5.11(b). Hence, the (quantization) noise in case of Figure 5.11(a) can be masked relatively easily than in case of Figure 5.11(b). Therefore, it is logical to assign sufficient number of bits to the subband with the lowest NMR. This criterion will be applied to all the subbands and until all the bits are exhausted. Typically, in audio coding standards an iterative procedure is employed that satisfies both the bit rate and global masking threshold ...