error when creating sound files

[mistify76]

I am trying to split audio from some video recordings I recently made, which are saved in MPEG-2 format. The video files are all around 4 GB in size, but some of them are quite long, as they were recorded at night, so there was a lot of compression. I have no problem splitting the audio other files created by the same recording system, but the longer ones seem to give Ulead a problem. I Create Sound File from .mpg video, and it goes through the whole process of rendering, then shows a "File reading error [17701:1:3]". The files show up in Windows Explorer, and appear to have content (i.e. they are several GB in size), but when I try to open them in Windows Media Player, RealPlayer, or signal processing software, they don't work.
Any thoughts on how to fix this would be greatly appreciated!
Thanks in advance!

[Black Lab]

Instead of trying to split the audio try making an audio file. With you clip in the timeline click on Share>Create Audio File. This will make a wav file from the clip. See if that works. If it does you can simply mute the audio on the video clip then insert this new wav file to the audio track.

[Ken Berry]

It would also be helpful to know what you intend to do with the split audio. Do you want to record it separately to a CD? Or do you want to process it separately through an audio editor to enhance certain parts or edit out unwanted noises and then reinsert it in your video project? Or do you just want it in the audio track of your VS project as opposed to remaining part of your main video clip(s)?

[mistify76]

The Share->Create Audio File is what I've been doing. Like I said, it works with some of the mpegs, but not others - the .wav file shows up in Windows Explorer, but nothing will read it. The video is part of my master's thesis, where we recorded video and audio of fish, and then I will be analyzing the acoustic portion to describe the calls (so will be using MATLAB and some other software to make some measurements on parts of the sound file). Basically, I need to be able to look through the sound files to find the fish sounds, and then go back to the mpegs to try and describe the behavior of the fish at the time they make sounds. Naturally, it would have been easier to just create separate audio and video files from the get-go, but our recorders only make mpegs, so I'm stuck having to generate wav files from the mpegs.
My only thought is that for some reason the wav files being generated are too big or long (because the mpegs are so compressed at night, they can be up to 10 hours long, so the wav file will be that long too). Might it be better to break up the original mpegs into smaller chunks?

[Black Lab]

That was going to be my next suggestion. If the big files are the ones giving you problems try cutting them into smaller bits before you create the audio file.

Fish make sounds?

[sjj1805]

Fish make sounds?

Fish Sounds