“LaBB-Cat is essentially a system that stores recordings of speech and their transcripts, and allows you to add layers of annotation on top of those transcripts. Once you’ve got the layers of annotation there, you can search for patterns and annotations across different layers. With the results of those searches, you can either explore the data directly with a LaBB-Cat or you can export the results to CSV files for further analysis. Other tools: you can also do batch acoustic measurement using integration with Pratt, and you can export the transcripts out in other other variety of other formats.”
There is a lot of functionality here, worth a look.
I worked through the first of the exercises in this Github repo (which are very good, I wonder why the site itself doesn’t link themThey do! There are some worksheets for self-directed exploration of LaBB-CAT’s functionality here which explain what to do.):
One of the goals of this tool seems to be the flexible import and export of many different formats — as you can see, the list is long:
(What is SALT?)
It also has some nice capabilities in terms of audio processing — you can carry out a search, and get an export of audio clips containing the match. Then you can export a TextGrid, and open up the lot in Praat.
There are a few things that are not immediately obvious, at least to me:
- Can you add new transcripts/audio to the system?
- Some of the pages are kind of opaque — what does keyness mean?
But this is more just me wondering, not a critique of the system. Check it out!