From User Perceptions to Technical Improvement: Enabling People Who Stutter to Better Use Speech Recognition
Colin Lea,Zifang Huang,Lauren Tooley,Jaya Narain,Dianna Yee,Panayiotis Georgiou,Tien Dung Tran,Jeffrey P. Bigham,Leah Findlater
DOI: https://doi.org/10.48550/arXiv.2302.09044
IF: 6.4588
2023-02-17
Human-Computer Interaction
Abstract:Consumer speech recognition systems do not work as well for many people with speech diferences, such as stuttering, relative to the rest of the general population. However, what is not clear is the degree to which these systems do not work, how they can be improved, or how much people want to use them. In this paper, we frst address these questions using results from a 61-person survey from people who stutter and fnd participants want to use speech recognition but are frequently cut of, misunderstood, or speech predictions do not represent intent. In a second study, where 91 people who stutter recorded voice assistant commands and dictation, we quantify how dysfuencies impede performance in a consumer-grade speech recognition system. Through three technical investigations, we demonstrate how many common errors can be prevented, resulting in a system that cuts utterances of 79.1% less often and improves word error rate from 25.4% to 9.9%.