Evaluating Low-Level Speech Features Against Human Perceptual Data
Published
2017-11-17
Caitlin Richter
,
Naomi H Feldman
,
Harini Salgado
,
Aren Jansen
Caitlin Richter
University of Pennsylvania
Naomi H Feldman
University of Maryland
Harini Salgado
Pomona College
Aren Jansen
Johns Hopkins University
Abstract
We introduce a method for measuring the correspondence between low-level speech features and human perception, using a cognitive model of speech perception implemented directly on speech recordings. We evaluate two speaker normalization techniques using this method and find that in both cases, speech features that are normalized across speakers predict human data better than unnormalized speech features, consistent with previous research. Results further reveal differences across normalization methods in how well each predicts human data. This work provides a new framework for evaluating low-level representations of speech on their match to human perception, and lays the groundwork for creating more ecologically valid models of speech perception.
PDF (presented at EMNLP 2017)
,