Role of head pose estimation in speech acquisition from distant microphones

Reverberant environments pose a challenge to speech acquisition from distant microphones. Approaches using microphone arrays have met with limited success. Recent research using audio-visual sensors for tasks such as speaker localization has shown improvement over traditional audio-only approaches. Using computer vision techniques we can estimate theorientationofthespeaker'sheadinadditiontothelocation of the speaker. In this paper we study the utility of using the head pose information for effective beamforming and clean speech acquisition from distant microphones. The improvementsinspeechrecognitionaccuracyrelativetothatofaclose talking microphone are presented and the results provide sufcient motivation for incorporating head pose information in beamforming techniques.

• Type: Conference
• Year: 2009
• Where: ICASSP
• Authors: Shankar T. Shivappa, Bhaskar D. Rao, Mohan M. Trivedi
• Keywords: Distant Microphones, Head Pose Information, ICASSP 2009, Signal Processing, Speech Acquisition