Hi, I am relatively new to using ELAN and I am trying to add the recognizer api. I use a mac, and I’m having trouble with the instructions in the readme.pdf document. Do I need to use Maven or Java or both to add the recognizer to ELAN? Does anyone have more detailed instructions (I’ve looked and can’t find instructions on how to successfully add the recognizer). I would like to somewhat automatically recognize different speakers. Thank you for any advice!
Hi,
Just to be sure: you only need the recognizer API in case you want to implement your own recognizer or want to integrate an existing recognizer that you have. Is that indeed what you plan to do, or do you just want to make use of existing recognizers that are maybe mentioned in the manual or additional documents?
-Han
Hi Han, thanks so much for your reply! I would like to make use of the existing recognizers mentioned in the manual so that I can recognize different speakers somewhat automatically/know the duration of each person’s turn. I was able to use the silence recognizer on ELAN but from what I read in the manual/online, I thought that the other recognizers needed to be added before they could be used.
I downloaded the zip file for the audio/video recognizer (https://www.mpi.nl/tools/elan/recognizer_api_V4.zip) but I’m having trouble getting it to work. From what I understand, I need to enter some of the files in Java/Maven? Thank you and apologies for my newbieness! I really appreciate the help!
Hi,
The existing recognizers are now part of the ELAN distribution and don’t need to be added separately anymore. These recognizers now run on a server and ELAN uploads the media file to that server for processing.
So, in the Recognizer tab in ELAN, where you selected the Silence recognizer, you should also see one named “Speaker diarization pipeline…”. You should be able to try this one (and the other recognizers in the list) without installation of additional extensions.
Ah ok! Thank you - I will try!