Maintaining Audio Quality in Cascaded Psychoacoustic Coding
A method is presented to reduce the distortion growth when a signal is repeatedly coded in a transmission chain of cascaded, lossy codecs. The method is suited for the MPEG Audio codecs. Psychoacoustic audio codecs achieve large compression ratios by adapting the signal precision to the human ear's acuity. Since that acuity is estimated from the input signal's characteristics, the coding noise in that signal, which is present due to previous encoding-decoding operation, is not accounted for. Consequently, cascaded coding poses the risk that the level of the coding distortion increases above the masked threshold, leading to audible artifacts. One cause of the noise increase is related to the cascading of synthesis and analysis filter in the decoder and encoder, respectively. In general, they will cause the quantization noise to smear out both in time and frequency. Only if the two filters are connected through a delay line causing the delay at the subband sample rate to be integer, this smear out is suppressed. Because the smear out is suppressed, this optimum delay can be recognized as the one where the signal power is minimum.
Click to purchase paper 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 $20 for non-members, $5 for AES members and is free for E-Library subscribers.