Download Latest Version
install-labbcat_20250430.jar (206.1 MB)
Get an email when there's a new version of LaBB-CAT
| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| README.txt | 2012-10-23 | 2.2 kB | |
| UpgradeONZEMinerToLaBBCAT.zip | 2012-07-08 | 40.7 MB | |
| Totals: 2 Items | 40.7 MB | 0 | |
==================================================== Why upgrade from ONZE Miner to LaBB-CAT? ==================================================== ONZE Miner is no longer actively supported, where LaBB-CAT development is ongoing. LaBB-CAT uses 'annotation graphs' to store data. As a result, things you can do now that you couldn't do with ONZE Miner include: * the possibility of having multiple sets of long-passage tags (in ONZE Miner, only the 'topic' layer is available for this, in LaBB-CAT you can have as many as you like, which you can annotated using Praat) * the possibility of having phrase/sentence annotations * the possibility of have syntactic parse annotations, and associated part of speech tags, generated by the Stanford Parser, which can be hand-corrected using a tree editor * the possibility of reconstructing syllable annotations from phones generated by HTK forced-alignment * processing is faster for most automatic annotation, and for searching the ability to upload Praat transcripts without them having to be structured in a strictly Transcriber-like manner, and ELAN transcripts are also now supported, as well as the new version of Transcriber - "TranscriberAG" * transcripts can be exported to ELAN Some other stuff that's been developed, more or less unrelated to annotation graphs, includes: * support for American English pronunciations loaded from the CMU Pronunciation Dictionary * support for accent-specific English pronunciations and other data from the Unisyn lexicon * automatic annotation of phonemes, words, and phrases based on regular expressions * annotation from CSV files (i.e. you can add a column to a search results export, fill it in with annotations, and upload the annotations into LaBB-CAT) * more options for selection of search results matches and related annotations, for export to CSV and other formats * HTK forced-alignment support on linux * if forced-alignment has been done, search results files can be reprocessed with Praat, to automatically extract formant measurements and other stuff on the basis of time-stamps in the CSV file * improved online help ...among other features.