Language-specific Effects on Automatic Speech Recognition Errors for World Englishes
Year: 2022
Type: other
Abstract: Despite recent advancements in automated speech recognition (ASR) technologies, reports of unequal performance across speakers of different demographic groups abound. At the same time, the focus on performance metrics such as the Word Error Rate (WER) in prior studies limit the specificity and scope of recommendations that can be offered for system engineering to overcome these challenges. The current study bridges this gap by investigating the performance of Otter’s automatic captioning system on native and non-native English speakers of different language background through a linguistic analysis of segment-level errors. By examining language-specific error profiles for vowels and consonants motivated by linguistic theory, we find that certain categories of errors can be predicted from the phonological structure of a speaker’s native language.
Source: Underline Science Inc.
DOI: 10.48448/wv0k-hr42
Authors: International Committee for Computational Linguistics 2022, Choe, June
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W6929286652