In this paper, an efficient speech recognition system is proposed for speaker-independent isolated digits (0 to 9). Using the Weighted MFCC (WMFCC), low computational overhead is achieved since only 13 weighted MFCC coefficients are used. In order to capture the trends of the extracted features, the local and global features are computed using the Improved Features for Dynamic Time Warping (IFDTW) algorithm. In this work, we propose to reduce the time complexity of the recognition system by time-scale modification using a SOLA-based technique and also by using a faster implementation of IFDTW. The experiments based on TI-Digits corpus demonstrate the effectiveness of proposed system giving higher recognition accuracy of 99.16% and performing about 22 times faster than conventional techniques.