DRIVE METHOD OF HIDDEN MARKOV MODEL PARAMETER RECOGNIZER USING CAR SPEED ADAPTIVE DATA BASE

摘要

PURPOSE: A drive method of HMM(hidden Markov model) parameter recognizer using car speed adaptive database is provided to extract voice recognition HMM parameters by car speed using a voice database containing car noises every car speed section. CONSTITUTION: A drive method of hidden Markov model parameter recognizer using car speed adaptive database comprises steps of: digitalizing input voice; dividing a voice section into a frame unit; extracting a characteristic vector every frame, and then quantizing the vector; measuring and classifying car speed; and comparing probabilities under the status of model parameter set, thereby outputting the recognition results.