Tree and filter optimization for audio compression in a wavelet-based perceptual audio coder

This paper outlines a new perceptual low bit rate audio coding scheme based on adapted wavelet representations. It claims wavelet tree and filter adaptation attending to a perceptual entropy-based method. To achieve such adaptive structure, a periodized wavelet packet transform is performed for each audio frame. After the transform, the encoder employs scalar adaptive quantization, controlled by an enhanced psycho-acoustic model, followed by entropy coding and bit packing. An adapted Huffman algorithm that assumes Laplacian distributions for wavelet coefficients is used for entropy coding. A relevant feature of our coder is the method to translate the psycho-acoustic information to the wavelet domain. It allows using orthonormal wavelets with any compact support. Experimental results indicate that our coder ensures ‘transparent’ coding of monophonic CD quality audio signals at bit rates lower than 64 kb/s for all the test signals. Key-Words: audio coding, wavelets, best basis selection, perceptual entropy, Laplacian distribution.