Parent-child
Up one level
Parent-child
I have just recently started studying how to work with elan (3.2.0). And I am having basic problems. And I can't explain these even in my mothertongue (Finnish) so in English this gets really difficult. I am hopeless with computers and computer programs.
First I tried to add a new tier (Word) and I tried to choose there already a parent (Speech) for the new one, but the programme wouldn't let me. Then I have tried to make one tier as a parent for another tier in the 'change the attributes -tool'. Well, when I try to make a change and I choose the Spch as parent, after making the choosing, in this same window the command box 'change' (or 'add') inactivates and I can't make the change valid. My only option in this point is close. After closing, none of the changes stay put.
I have also tried to work with the tokenizer and there I choose for source tier this Spch. And then I need to choose destination. There are no options. Therefore I choose this add a new tier. And while working in the window that next opens, I still have the same problem, that the program won't let me choose the parent tier. Or I can choose it, but I can't click on the add, because it becomes inactive.
I hope this explanation gives some idea of my problem. And I had big problems in understanding the manual. I figured the first steps via trial and error -method and not with the manual.
Minna
Re: Parent-child
I managed to figure out, how to make this parent-child tiering.
Now I have a new problem.
When I add a linguistic type for a tier, the program won't let me change the language type as an utterance for the first tier that I made. Before trying to do this I had first added 2 linguistic types I wanted (utterance and words). But for the following tier I can choose the linguistic type from those options I have made.
What do I do wrong?
Minna
Re: Re: Parent-child
In general a tier without a parent (a top level tier, Speech in your example) must have a linguistic type with stereotype "None". A dependent tier (like Word in your example) must have a linguistic type with a stereotype other than "None", e.g. Symbolic Subdivision. See http://www.lat-mpi.eu/tools/elan/manual/ch04s01.html.
If you have annotations on the top level tier Speech and you have a depending tier Word with a "Symbolic Subdivision" linguistic type, you can use the tokenizer with Speech as source and Word as destination.
It isn't always possible to change the linguistic type of an existing tier, especially if you want to choose a linguistic type with a different "Stereotype". The linguistic type puts constraints on annotations of a tier and annotations could be destroyed when changing the linguistic type. That is why the application tries to prevent that.
It should always be possible to change the "Default language" of a tier.
Han
Release of ELAN 4.3.0
