Feature #3049
UDPipe annotation engine, tokenizer
| Status: | New | Start date: | 04/09/2021 | |
|---|---|---|---|---|
| Priority: | Normal | Due date: | ||
| Assignee: | - | % Done: | 0% |
|
| Category: | Annotation | Spent time: | - | |
| Target version: | TXM - Eltec 1.0 |
Description
Integrate the UDPipe annotation engine in TXM like TreeTagger is.
If available, The UDPipe annotation engine can be selected in the import form instead of TreeTagger (in the Language section).
To work properly, the UD tagger needs the appropriate tokenization.
During the import process, if the UDPipe annotation engine is selected, TXM will not use its tokenisation rules but the UD pipe model tokenisation.
Related issues