This Is Me: Using Ambient Voice Patterns for In-Car Positioning
With the range of services that can be accessed inside a car constantly increasing, so are the opportunities for personalizing the experience for both driver and other passengers. A main challenge however is to find out who is sitting where without asking explicitly. The solution presented in this paper combines two sources of information in a novel way: Ambient speech and mobile personal devices. The approach offers improved privacy by putting the user in control, and it does not require specialized positioning technologies such as RFID. In a data-driven evaluation, we confirm that the accuracy is sufficient to support a ten-speaker scenario in practice.
KeywordsPositioning Speaker Recognition Automotive HMI User Modeling Personalization
Unable to display preview. Download preview PDF.
- 1.Brouwer, R.F.T., Hoedemaeker, M., Neerincx, M.A.: Adaptive interfaces in driving. In: HCI, vol. (16), pp. 13–19 (2009)Google Scholar
- 2.Corkill, D.D.: Blackboard systems. AI Expert 6, 40–47 (1991)Google Scholar
- 3.Feld, M., Müller, C.: An Integrated Development Environment for Speech-Based Classification. In: Proceedings of the 13th International Conference “Speech and Computer” SPECOM 2009, St. Petersburg, Russia, pp. 443–447 (June 2009)Google Scholar
- 4.Hayashi, H., Tsubaki, T., Ogawa, T., Shimizu, M.: Asset tracking system using long-life active rfid tags. NTT Technical Review 1(9), 19–26 (2003)Google Scholar
- 5.Reynolds, D.A., Quatieri, T.F., Dunn, R.B.: Speaker verification using adapted gaussian mixture models. In: Digital Signal Processing, p. 2000 (2000)Google Scholar
- 6.Schwartz, T., Stahl, C., Baus, J., Wahlster, W.: Seamless Resource-Adaptive Navigation. In: Resource-Adaptive Cognitive Processes, pp. 239–265. Cognitive Technologies, Springer (2010)Google Scholar