home edit page issue tracker

This page pertains to UD version 2.

Metadata harmonisation: align spoken-data fields with UniDive naming conventions

Back to South_Levantine_Arabic MADAR · Back to index

Repo: https://github.com/UniversalDependencies/UD_South_Levantine_Arabic-MADAR

Cross-posting from the UniDive WG1 T1.5 (spoken language guidelines) metadata harmonisation review. We compared UD_South_Levantine_Arabic-MADAR’s current CoNLL-U metadata against the proposed naming conventions (see also the full treebank status table). This is a suggestion for maintainers to review - feel free to push back on anything that doesn’t fit the corpus. The comparison was carried out semi-automatically with the help of Claude (Anthropic); errors or misunderstandings are possible, so please double-check anything unclear.

1. Is the spoken portion identifiable?

genre lists spoken social, but the README indicates the 100 sentences are actually written: they’re manually translated (not transcribed) short conversational tourism-related texts from the MADAR Parallel Corpus, itself derived from the written Basic Traveling Expression Corpus (BTEC). We couldn’t find any indication that a sentence is a transcription of actual speech.

Suggestion: Could you confirm whether any sentences are actually spoken, or should spoken be dropped from genre?

Implementation notes


This issue was prepared as part of the UniDive WG1 T1.5 spoken language guidelines effort. Happy to help implement these changes ourselves if that’s easier than doing it on your end - just let us know.