SPEECH ENHANCEMENT FOR A CAR ENVIRONMENT SUPPORT BY A FIRST-ORDER DIFFERENTIAL MICROPHONE

摘要

Speech recognition is one of the key technologies to produce voice-control devices. However, the presence of different sources of degradation is an ubiquitous problem. This paper presents a robust microphone array processing technique to enhance speech under the influence of noise and reverberation in an automobile environment. The proposed structure combines a simple two-microphone First Order Differential Null Beamformer with Spectral Subtraction techniques. The paper also includes an evaluation of the performance of the algorithm in a real car environment. The results show a noticeable reduction in word error rates when the enhancement front-end is applied to a standard speech recognizer.