Vocal resonance as a passive biometric

We anticipate the advent of body-area networks of pervasive wearable devices, whether for health monitoring, personal assistance, entertainment, or home automation. In our vision, the user can simply wear the desired set of devices, and they “just work”; no configuration is needed, and yet they discover each other, recognize that they are on the same body, configure a secure communications channel, and identify the user to which they are attached. This paper addresses a method to achieve the latter, that is, for a wearable device to identify the wearer, allowing sensor data to be properly labeled or personalized behavior to be properly achieved. We use vocal resonance, that is, the sound of the person’s voice as it travels through the person’s body. By collecting voice samples from a small wearable microphone, our method allows the device to determine whether (a) the speaker is indeed the expected person, and (b) the microphone device is physically on the speaker’s body. We collected data from 25 subjects, demonstrate the feasibility of a prototype, and show that our method works with 77% accuracy when a threshold is chosen a priori.

[1]  Douglas A. Reynolds,et al.  Speaker identification and verification using Gaussian mixture speaker models , 1995, Speech Commun..

[2]  David Kotz,et al.  Recognizing whether sensors are on the same body , 2011, Pervasive Mob. Comput..

[3]  Sharath Pankanti,et al.  Guide to Biometrics , 2003, Springer Professional Computing.

[4]  Douglas D. O'Shaughnessy,et al.  Speech communication : human and machine , 1987 .

[5]  Jie Liu,et al.  SpeakerSense: Energy Efficient Unobtrusive Speaker Identification on Mobile Phones , 2011, Pervasive.

[6]  Zhigang Liu,et al.  Darwin phones: the evolution of sensing and inference on mobile phones , 2010, MobiSys '10.

[7]  P. Mermelstein,et al.  Distance measures for speech recognition, psychological and instrumental , 1976 .

[8]  E. B. Newman,et al.  A Scale for the Measurement of the Psychological Magnitude Pitch , 1937 .

[9]  David Kotz,et al.  Privacy in mobile technology for personal healthcare , 2012, CSUR.

[10]  David V. Anderson,et al.  Cooperative analog-digital signal processing , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[11]  D.M. Mount,et al.  An Efficient k-Means Clustering Algorithm: Analysis and Implementation , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Shingo Kuroiwa,et al.  Combination method of bone-conduction speech and air-conduction speech for speaker recognition , 2008, INTERSPEECH.

[13]  Douglas A. Reynolds,et al.  A Tutorial on Text-Independent Speaker Verification , 2004, EURASIP J. Adv. Signal Process..

[14]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[15]  Bayya Yegnanarayana,et al.  Throat microphone signal for speaker recognition , 2004, INTERSPEECH.

[16]  Douglas A. Reynolds,et al.  Speaker Verification Using Adapted Gaussian Mixture Models , 2000, Digit. Signal Process..

[17]  Patrick Traynor,et al.  (sp)iPhone: decoding vibrations from nearby keyboards using mobile phone accelerometers , 2011, CCS '11.

[18]  Douglas A. Reynolds,et al.  Comparison of background normalization methods for text-independent speaker verification , 1997, EUROSPEECH.

[19]  G. Fairbanks Voice and articulation drillbook , 1960 .