Seung Kwon Beack

Blind source separation and Kalman filter-based speech enhancement in a car environment

In this paper, we present a speech enhancement method as a pre-processor in a car environment. For the friendly and safe use of the voice commander in a running car, non-stationary audio signals such as music and non-candidate speech should be reduced. Our technique is a two microphone-based one. It consists of two parts, i.e., blind source separation (BSS) and Kalman filtering. Firstly, BSS is operated as a spatial filter to deal with non-stationary signals and then car noise is reduced by Kalman filtering. The algorithm performance is tested with a speech recognizer. The results show that our algorithm can be very helpful for the voice commander in a car environment.

Additional data packetizing method for providing multichannel audio service on T-DMB environment

Terrestrial digital multimedia broadcasting(T-DMB) is one of mobile broadcasting services, and the commercial service was started in December 2005 in Korea. The performance targets of T-DMB are providing VCD(video CD) quality video and FM radio quality audio. In recent years, the researches for providing high quality video or audio service on T-DMB environments have been being carried out. To provide high-quality video or audio service, some additional data should be transmitted to the receiver as well as T-DMB video and audio data. Since the data rate for one T-DMB program is low, it is important to transmit the additional data at a low bit rate. In this paper, we propose a packetizing method for efficient transmission of the additional data to provide multichannel audio service on T-DMB environment.