In the research on speech recognition of small vocabulary and isolated words, the recognition rate of the words with fast (or slow) rate or high (or low) volume is degraded. Therefore, it is necessary to adjust their rate or volume. Based on the linear prediction analysis technique, the paper propose a novel adjusting methodology, whereby the speech signal to be adjusted is first transformed by the antisolution to the glottal pulse to have their rate and volume adjusted. Through the sound channel system, these adjusted glottal pulses are transformed to the adjusted speech with proper rate and volume. Simulation results show that the method is practically feasible and able to improve the recognition rate of the speech recognition system for the words with fast (or slow) rate or high (or low) volume.