Speech recognition device, speech enhancement device, speech recognition method, speech enhancement method and navigation system

摘要

An apparatus includes a plurality of noise suppression units (3) that perform respective noise suppression processes using different methods of speech data with noises input to them. The apparatus further includes: a voice recognition unit (4) that performs voice recognition on audio data generated by suppressing a noise signal in the voice data with noise; a prediction unit (2) that predicts from acoustic characteristic quantities of the input speech data with noises speech recognition rates to be provided when the noise suppression processes on the speech data with noises are respectively performed by the plurality of noise suppression units (3); and a suppression method selection unit (2) that selects a noise suppression unit (3) that performs a noise suppression process on the speech data with noise from among the plurality of noise suppression units based on the predicted speech recognition rates.