This paper presents a continuous VQ clustering (CVQC) algorithm for realtime speech recognition, which incorporates the temporal information of speech into both training and recognition processes. In comparison with the conventional DTW and VQ methods, this new algorithm delivers faster training and recognition speed and smaller codebook size while still retains merits of both. Realtime implementation is emphasized in the design of sophisticated algorithms. And a custom available voice controlled computer command input system based on CVQC is also introduced.