I’m a new user and am combing through the quick guides, but I thought it best to ask it here. I have 15-16 hour .wav files, and I am hoping to find a measure of conversational time (speech) within that sample. Is there a way to find a measure of this in ELAN? Can I find a cumulative “talking time” in a large .wav file? I also have the same file in .trs (a transcription file), and I’m currently trying to import one to play around with it.
Depends a bit on what you expect; if you have transcribed the wav files it is easy to get an overview in the annotation statistics. Either per file via View->Annotation Statistics or for multiple files via File->Multiple File Processing->Statistics for Multiple Files. It depends a bit on how the speech has been annotated (on a single tier per audio file, or on multiple tiers in case of multiple speakers etc.) whether you have to add up the total durations of multiple tiers or will find the total duration directly as it is on one tier.
If you have no annotations at all (and therefore no segmentation into speech/non speech segments) you might want to try automatic segmentation by one of the recognizers in the Audio Recognizer tab?
-Han