Research and developments of a multi‐modal MIR engine for commercial applications in East Asia1

Abstract

This article describes the research and development of an efficient Music Information Retrieval (MIR) engine that is embedded in a karaoke software package targeted for Asian people's need of music retrieval. The MIR engine has a multi-modal interface that allows queries by singing, humming, tapping, speaking, and writing. In particular, we discuss the design philosophy, technical barriers, and performance evaluation of such an engine, as well as its current and potential commercial applications. Feedbacks and feature requests from users, which greatly influence our future work, are also addressed.

An Evaluation of Feature Extraction for Query-by-Content Audio Information Retrieval

“…Choice of low-level audio feature is a very important stage in Contentbased Music Information Retrieval (CBMIR). Many studies have shown the need of feature choice, such as Mel-Frequency Cepstral Coefficients (MFCC) [1][2][3], Pitch [4][5][6], Short Time Fourier Transform (STFT) [7][8][9] and so on, that have been proposed for the extraction of salient information and diminish redundancy. Unfortunately, little work focuses on analyzing music signal properties and comparing the performance of retrieval using these features.…”