Microsoft is committed to developing a new voice input interface that allows users to speak and record without the presence of sound. The study was conducted by Microsoft Research and was introduced in ACM CHI 2018. This technique is called SilentVoice to communicate by recording the sounds of breathing, which allows a whisper-like sound to be enough for the microphone to record the voice without disturbing the people around. In addition, the module will filter the surrounding voice, so users can capture clear voice even if there is outside interference.
SilentVoice is a new voice input interface device that leverages the voice-based Natural User Interface (NUI) in everyday life.
The proposed "progressive speech" method is capable of placing the microphone very close to the front of the mouth without being affected by noise, capturing very soft speech with a good S/N ratio. It enables ultra-small (less than 39dB(A)) voice leakage, allowing voice input without annoying people around the public and mobile, as well as in the office and home. (At last I won’t bother people to use TNT!)
By measuring the direction of the airflow, SilentVoice can easily separate external sounds from normal speech with an accuracy of 98.8% and no need to read the activation words before the voice communication starts. It can also be used with a voice activation system with a specially trained speech recognizer. The resulting word error rate (WERs) is 1.8% (speaker dependent condition), 7.0% (speaker unconditional), including 85 Command sentences, which means that natural voices like whispers can also be used for real-time voice communication.
You can view the full presentation at the ACM CHI Computing System Conference:https://youtu.be/9EV1mEtVfuM
The technology is still in the research stage, but it will definitely help those who prefer to use voice commands but prefer to work without disturbing the people around them.