PARAMETRIC AUDIO CODER BASED ON SPARSE APPROXIMATION WITH FRAME-BASED PSYCHOACOUSTIC OPTIMIZED WAVELET PACKET DICTIONARY

Abstract

Time-frequency signal transform is considered. Advantages of joint time-frequency analysis over traditional separate analysis in time or frequency domains are shown. Matching pursuit algorithm and modified psychoacoustic matching pursuit describes. Universal audiocoder scheme based on sparse approximation with frame-based psychoacoustic optimized wavelet packet dictionary proposed.

Keywords

• частотно-временные преобразования
• пакет дискретного вейвлет преобразования
• согласованная подгонка
• психоакустическая модель
• аудиокодер