SDITDec 14, 2015

Trigonometric dictionary based codec for music compression with high quality recovery

arXiv:1512.04243v12 citations
Originality Incremental advance
AI Analysis

This addresses high-quality music compression for audio applications, but appears incremental as it builds on existing transform compression methods.

The authors tackled music compression by proposing a transform lossy codec using a redundant trigonometric dictionary and pursuit strategy for sparse approximation, achieving remarkable improvements in compression performance compared to OGG Vorbis while maintaining identical decompressed signal quality.

A codec for compression of music signals is proposed. The method belongs to the class of transform lossy compression. It is conceived to be applied in the high quality recovery range though. The transformation, endowing the codec with its distinctive feature, relies on the ability to construct high quality sparse approximation of music signals. This is achieved by a redundant trigonometric dictionary and a dedicated pursuit strategy. The potential of the approach is illustrated by comparison with the OGG Vorbis format, on a sample consisting of clips of melodic music. The comparison evidences remarkable improvements in compression performance for the identical quality of the decompressed signal.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes