SPEECH RECOGNITION DEVICE

摘要

PROBLEM TO BE SOLVED: To provide a speech recognition device capable of improving efficiency of speech learning. SOLUTION: The speech recognition device is a device which performs speech learning for recognizing speech by using an input speech signal. The speech recognition device generates a plurality of signal sequences by analyzing a signal from different time for the input speech signal and performs learning for each of the plurality of signal sequences. The plurality of signal sequences are generated by extracting a feature amount for each frame in which the frame is shifted with a predetermined frame period from the different time for the input speech signal. The different time is shorter time than the frame period for the input speech signal. COPYRIGHT: (C)2009,JPO&INPIT