Datasets:

Modalities:
Text
Formats:
parquet
DOI:
Libraries:
Datasets
pandas
License:

Suggestions for improvement

#2
by asounimelb - opened

Thank you for compiling this valuable dataset. To improve usability, I suggest the following:

  • Add a new field for ISO 639 language codes.
  • Remove the unsegmented version when a record already contains both segmented and unsegmented glossing, since segmentation marks can be removed easily.

Sign up or log in to comment