Hello,
I use the Export Multiple Files as → Annotation Overlaps Information function in ELAN often for my work investigating signed languages. I like this function because it allows one to choose a reference tier and then export everything it overlaps with, even empty annotations or multiple annotations.
However, the resulting format of the export is difficult, and so far requires a lot of manual re-formatting, and I wonder if something about this could be done?
The exported data is organized by row and “chunked” by each file (meaning there is a empty row above and below each group of hits. The first column is the Begin Time… but in this column, above each “group” is actually the file name with its full path. This repeats for each file.
Previously, I have had to manually copy the file name over to a new column, copy it down to only the hits that it is associated with, and repeat this over and over.
I am now wondering if something can be edited from the ELAN side, so that in the export the file name associated with each hit can be added in its own column?
I have attached a picture: The top of the spreadsheet is what I would like….below is what ELAN produces now. 
Or, if anyone knows who to address this in Excel or R to make the re-formatting more automatic?
I thank you in advance for your time and any advice.
Kind regards, Lindsay Ferrara
Hi Lindsay,
The Export Multiple Files as → Time Co-reference Tab-delimited Text… introduced in ELAN 7.1 was developed specifically in response to the limitations of the existing Annotation Overlaps Information export — and it sounds like it may be exactly what you’re looking for.
In essence, it is a streamlined version of that export, with the following key differences:
- One field per tier (the annotation value only), rather than the eleven fields generated by Annotation Overlaps Information. That said, if you were actually making use of some of those other ten fields in your analysis, this option may not be the right fit — but if the annotation value was all you needed, this should work much better.
- Filename included as a dedicated field (i.e. a column in the CSV), rather than appearing as part of a file-path header. This makes it straightforward to uniquely identify each data point by filename and timecode — without any post-export scripting.
- Configurable concatenation separator — the export dialog lets you specify your preferred separator symbol, rather than defaulting to a comma. This is useful when commas or other common punctuation may already appear within annotation values.
The underlying overlap criteria are identical to those used by Annotation Overlaps Information — only the output has been simplified.
I hope this export option suits your workflow.
I have tried this – and it works, and is amazing. Thank you so much for your quick and helpful reply! Kind regards, Lindsay
Glad to hear it worked well for you. Thank you for the feedback.
