symbolic segmentation
Up one level
symbolic segmentation
I am new to ELAN and wondering what is the most efficient way to created (non-time aligned) symbolic segmentations. Unfortunately, the segmentation is done on semantic criteria and therefore cannot be done automatically.<br><br>Also, is it possible to do annotations with ELAN that are non-continuous e.g. Annotating evaluative language in a text, where what I want to annotate is just the evaluative lexis and not the other words in the text. Do I have to chunk the entire text into words first?<br><br>And finally, I cannot work out how to 'merge' annotated units that are next to eachother.<br><br>Many thanks!<br><br>Michele<br>
Re: symbolic segmentation
If the symbolic segmentation cannot be done automatically by the tokenizer the only other option is to create it manually. If you have created the Symbolic Subdivision depending tier, start with double clicking in the area below an annotation on the parent tier. Once you have this first annotation, select it (make it active) and choose Annotation -> New Annotation After/Before (Alt+Shift+N or Ctrl+Shift+N). And so forth.
It is possible to create non-continuous annotations on a tier of type "Included In". Create a linguistic type with Stereotype Included In and then create a depending tier using this type. On tiers without parent it is of course always possible to create non-continuous annotations.
Currently there is only the option to merge 2 transcription files. Merging adjacent annotations is not yet implemented (but high on the to do list). At the moment the only way is to copy/paste the text from one annotation to another.
Han
Release of ELAN 4.3.0
