Comparative evaluation of automated scoring of syntactic competence of non-native speakers

作者：

Highlights：

• Comparison of various grammar features in context of automated speech scoring.

• Features related to shorter text spans show higher correlations with human scores.

• Adding grammar features to baseline model improves correlation with human scores.

• Computing features on concatenation of spoken responses improves performance.

摘要

•Comparison of various grammar features in context of automated speech scoring.•Features related to shorter text spans show higher correlations with human scores.•Adding grammar features to baseline model improves correlation with human scores.•Computing features on concatenation of spoken responses improves performance.

论文关键词：Automated scoring,Automated speech recognition,English language assessment

论文评审过程：Received 5 July 2016, Revised 16 January 2017, Accepted 29 January 2017, Available online 31 January 2017, Version of Record 9 September 2017.

论文官网地址：https://doi.org/10.1016/j.chb.2017.01.060