Published October 29, 2026 | Version 1.1

Analysis code and data for "Automatic cross-lingual annotation of clause-based utterance fluency in L2 speech"

  • 1. ROR icon Institute of Science Tokyo

Description

Analysis code, reference annotation and results for the article "Automatic cross-lingual annotation of clause-based utterance fluency in L2 speech" (Research Methods in Applied Linguistics). With the two corpora, these files reproduce every RQ1-RQ3 number of the article.

The recordings come from ALLSSTAR (40 L2 English story retellings) and I-JAS (40 L2 Japanese story retellings). The original ALLSSTAR and I-JAS data (audio and transcripts) are not included; they are available from the corpus providers under their own licences. In the TextGrids, ASR output and annotation files, the speakers' words are replaced by pointers. restore_corpus_text.py puts the words back from the user's own copy of the corpora. See README.md.

Files

analysis-code-and-data-v1.1.zip

Files (7.9 MB)

Name Size
md5:a123337cb8fa8822dfae452db9397f04
7.9 MB Preview Download

Additional details

Dates

Submitted
2026-10-29