Source modeling approaches to enhanced decoding in lossy audio compression and communication.

摘要

Mathematical modeling of audio signals plays a crucial role in a wide range of audio engineering problems including audio compression, modification and synthesis. The term "signal modeling" refers to the task of describing a signal's fundamental behavior by means of an appropriate model and its parametric information. As source modeling provides accurate and compact representation, it has been utilized in digital audio compression to achieve better reproduction quality at lower bit-rate. Beside the compression aspect, source modeling of audio signal enables signal analysis and characterization as well as the related task of estimation of missing components in source signal.;This dissertation presents effective utilization of source modeling in low bit-rate audio compression and communication. A high frequency regeneration (HFR) scheme is developed in the context of high-efficiency advanced audio coding (HE-AAC), which employs sinusoids plus noise model to accurately estimate the high frequency components of source signal. Utilization of source modeling is next extended to audio frame loss concealment (FLC), a demanding technique in audio communication to mitigate an adverse impact of transmission error on decoder reconstruction quality.