Bioinspired Auditory SoundLocalisation forImproving theSignal toNoise Ratio ofSocially Interactive Robots

Stefan Wermter

Abstract

Inthispaperwedescribe abioinspired hy- bridarchitecture foracoustic soundsourcelocalisation andtracking toincrease thesignal tonoise ratio (SNR) betweenspeaker andbackground sources fora socially interactive robot's speech recogniser system. Themodel presented incorporates theuseofInteraural TimeDiffer- enceforazimuth estimation andRecurrent NeuralNet- worksfortrajectory prediction. Theresults arethenpre- sented showing thedifference intheSNRofalocalised and non-localised speaker source, inaddition topresenting the recognition ratesbetweena localised andnon-localised speaker source. Fromtheresults presented inthis paperit canbeseenthatbyorientating towards thesoundsource ofinterest therecognition rates ofthatsource canbein- creased. IndexTerms- SoundSourceLocalisation, Signal to NoiseRatio, Speech Recognition, Sociable Interactive Ro- bots, HumanRobotInteraction.
BibTeX
@inproceedings{wermter2006bioinspired,
  author = {Stefan Wermter},
  title = {{Bioinspired Auditory SoundLocalisation forImproving theSignal toNoise Ratio ofSocially Interactive Robots}},
  booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems},
  year = {2006},
  url = {https://openreview.net/forum?id=jyf8YJfKiU},
}

Loading related content...

Figure