Summary of the paper
| Title | The Influence of the Utterance Length on the Recognition of Aged Voices |
|---|---|
| Authors | Alexander Schmitt, Tim Polzehl, Wolfgang Minker and Jackson Liscombe |
| Abstract | This paper addresses the recognition of elderly callers based on short andnarrow-band utterances, which are typical for Interactive VoiceResponse (IVR) systems. Our study is based on 2308 short utterances from adeployed IVR application. We show that features such asspeaking rate, jitter and shimmer that are considered as most meaningful onesfor determining elderly users underperform when used inthe IVR context while pitch and intensity features seem to gain importance. Wefurther demonstrate the influence of the utterance lengthon the classifier’s performance: for both humans and classifier, thedistinction between aged and non-aged voices becomes increasinglydifficult the shorter the utterances get. Our setup based on a Support VectorMachine (SVM) with linear kernel reaches a comparablypoor performance of 58% accuracy, which can be attributed to an averageutterance length of only 1.6 seconds. The automatic distinctionbetween aged and non-aged utterances drops to random when the utterance lengthfalls below 1.2 seconds. |
| Language | Person Identification |
| Topics | Statistical and machine learning methods, Emotion Recognition/Generation, Person Identification |
Full paper  |
The Influence of the Utterance Length on the Recognition of Aged Voices |
| Bibtex | @InProceedings{SCHMITT10.159, author = {Alexander Schmitt, Tim Polzehl, Wolfgang Minker and Jackson Liscombe}, title = {The Influence of the Utterance Length on the Recognition of Aged Voices}, booktitle = {Proceedings of the Seventh conference on International Language Resources and Evaluation (LREC'10)}, year = {2010}, month = {may}, date = {19-21}, address = {Valletta, Malta}, editor = {Nicoletta Calzolari (Conference Chair), Khalid Choukri, Bente Maegaard, Joseph Mariani, Jan Odjik, Stelios Piperidis, Mike Rosner, Daniel Tapias}, publisher = {European Language Resources Association (ELRA)}, isbn = {2-9517408-6-7}, language = {english} } |