VOICE RECOGNITION METHOD AND DEVICE

摘要

A voice recognition method, comprising: acquiring voice data (S100); according to the voice data, acquiring a first confidence value (S101); according to the voice data, acquiring a noise scenario (S102); according to the first confidence value, acquiring a second confidence value corresponding to the noise scenario (S103); and if the second confidence value is greater than or equal to a pre-stored confidence threshold, processing the voice data (S104). And a device. This method and device for flexibly adjusting a confidence value according to a noise scenario greatly improves the voice recognition rate in a noisy environment.