Sang Bae Chon

A bit reduction algorithm for Spectral Band Replication based on human auditory characteristics

Spectral band replication (SBR) is an effective tool for low bitrate audio codecs such as HE-AAC and enhanced aacPlus. We propose a new algorithm for the bitrate reduction of SBR, which modifies the envelope data coding of SBR considering the threshold in quiet and the masking threshold. Experimental results show that the proposed algorithm significantly reduces the bitrate compared to the conventional SBR with no degradation in perceived sound quality.

Sound field processing system using grouped reflections algorithm for home theater systems

In this paper, a new novel sound field processing system is proposed. This sound field processing system assists listeners in perceiving audible characteristics of a given movie theater or hall, especially with regard to distance and spatial impression. To extend sound space artificially, conventional algorithms reproduce reflected sounds based on the panning law. However, even if this panning law could extend perceived listening space, it accompanies sound coloration, which reduces sound quality. Here, grouped reflections algorithm (GRA) for sound field processing is proposed to reproduce sound fields without sound coloration. Moreover, the algorithm provides higher efficiency, reducing memory requirements of the sound field processing system.

Linear-scale perceptual feature extraction for Speech Bandwidth Extensions

This paper presents a new method to extract linear-scale perceptual feature as a subsitute of MFCCs for highband (3.4kHz∼) in Speech Bandwidth Extensions(BWE). The feature extraction method is based on the mel-scale constrained Nonnegative Matrix Factorization(NMF), which decompose linear-scale log spectrum into a linear combination of mel-scale latent variables. While MFCCs parametrization contains non-invertible procedures, suggested feature is represented in linear-scale and proper to recover the highband time-domain speech. Experiment results report that suggested feature shows better instrumental performance with narrowband MFCCs than real cepstrum without additional computation.

A novel audio stream segmentation method for audio signal discrimination

In this letter, an online segmentation algorithm for audio signal discrimination is presented. By detecting abrupt changes in audio signal features and decimating them followed by strength thresholding, segment boundaries of the audio stream are obtained. The resulting segment boundaries provide efficiency and accuracy for the classification stage of audio signal discrimination system.

Novel LMS algorithms based on status categorization

Least Mean Square (LMS) is an effective adaptive filtering algorithm with advantages of robustness and simplicity. In this paper, we propose two new algorithms, Categorized Variable Step Size LMS (CVSSLMS) and Combined CVSSLMS (CCVSSLMS), based on the categorization of filter status. The step sizes of the proposed algorithms are dynamically updated by optimization for each state. Experiment results show that the proposed algorithms outperform conventional LMS algorithms in both simplicity and robustness.