PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
May 14, 2026The Journal of the Acoustical Society of America0 citations

From subjective to scalable: Using automated proficiency scoring in L2 speech research

View Full Paper
SPS. Park

Key Points

  • This study aims to determine if automated assessments of speaking proficiency can replace human judgment in L2 research.
  • Employs multi-input, multi-output deep learning models to evaluate speaking proficiency.
  • Compares machine-generated scores with human ratings of speech quality.
  • Analyzes the relationship between speaking proficiency and segment production.
  • The automated system accurately reflects native listener evaluations of L2 speech.
  • It captures variability in L2 production similar to human-rated assessments.
  • Findings indicate a potential reduction in reliance on human annotations for proficiency scoring.

Abstract

Speaking proficiency is a critical factor in second language (L2) research, since it affects segmental and suprasegmental aspects of speech. However, obtaining proficiency measures often relies on standardized tests or subjective human ratings, both of which can be costly, time-consuming, or unavailable. This study investigates whether an automatic speaking proficiency assessment system can serve as a reliable substitute for human ratings in L2 research. Using multi-input, multi-output deep learning models, we examine whether the relationship between speaking proficiency and segment production, previously demonstrated with human-rated proficiency, can be replicated with machine-predicted scores. Results show that the automated system effectively mirrors native listener judgments, accurately capturing variability in L2 production. These findings suggest that automated proficiency assessments can reduce reliance on human annotation, offering a scalable and efficient tool for streamlining L2 experimental workflows.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

S. Park (2025) studied this question.

synapsesocial.com/papers/6a0567a8a550a87e60a1fc54https://doi.org/10.1121/10.0041528
Ask AI
Helpful
Bookmark
Share
View Full Paper