Many devices that include built-in microphone(s) are used in windy situations. Wind noise degrades the quality of audio detected by the microphone(s), causes microphone signal saturation at high wind speeds, causes nonlinear acoustic echo, and reduces the performance of acoustic echo cancellation (AEC). Applications such as voice‐trigger, automatic speech recognition (ASR), and voice over internet protocol (VoIP) communication are negatively impacted by such degradation.

This disclosure describes cost‐effective and robust techniques to detect and reduce wind noise. The described techniques deliver optimum removal and detection results by processing the audio signal in a holographic way by dealing with all related domains including time, frequency, and 3D space. This approach can improve the audio detection performance of any device that incorporates the techniques and can thereby improve the user experience of various applications such as voice-trigger, speech recognition, voice communication, event detection, etc. even on devices that have limited computational capability.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.