Interspeech'2005 - Eurospeech

Lisbon, Portugal
September 4-8, 2005

Remodeling of the Sensor for Non-Audible Murmur (NAM)

Yoshitaka Nakajima (1), Hideki Kashioka (2), Kiyohiro Shikano (2), Nick Campbell (2)

(1) ATR-HIS, Japan; (2) Nara Institute of Science and Technology, Japan

We developed the next generation of skin-attachment sensors for sampling NAM (Non-Audible Murmur) signals by using soft silicone, which has an acoustic impedance close to that of human flesh, as the prime medium of vibration. With new NAM microphones we could sample expanded target voice signal, suppressing air conduction noise signal to low by the experiment of synchronous stereo sampling of air and flesh conduction voices at the same gain. The bandwidth of the soft silicone type NAM microphone has improved and we obtain a much higher accuracy of both NAM and BTOS (Body Transmitted Ordinary Speech) recognition compared with the stethoscopic type. Aural comprehension test showed that accuracy of catching sentences of NAM and BTOS has improved with soft silicon type NAM microphone almost as high as that of air conduction voices. However, the extremely low accuracy of meaningless words is a problem to be solved for developing a "Non-Voice Phone".

Full Paper

Bibliographic reference.  Nakajima, Yoshitaka / Kashioka, Hideki / Shikano, Kiyohiro / Campbell, Nick (2005): "Remodeling of the sensor for non-audible murmur (NAM)", In INTERSPEECH-2005, 389-392.