A 30-minute 1920×1080 WMV is very large, like around 50MB.  We want it to be smaller before we publish it.  One way to make it smaller is to simply scale it down, but when the entire 1920×1080 canvas is shrunk down to a reasonable size, text which was perfectly legible before becomes unreadable.  Since the text is usually a crucial part of the demo, that’s unacceptable.

Usually, the presenter isn’t using the whole screen at any given time.  Rather, he’ll be focused on the content in one particular window for an extended period of time, and that window often takes up only around half of the screen.  Therefore, cropping the WMV to the active area is a possible way to the video file smaller without sacrificing legibility.  There’s just one problem — what happens when the presenter jumps around from window to window during the demo?

There are various expensive commercial video editors which will let you crop different segments of the video differently and then glom them all together into one video at the end.  However, I don’t like using commercial software when I can do it for free (this is probably about half on principle and half out of stubbornness).  Therefore, I set out to figure out how to do this with free tools.  The result is a script called wmv-to-panned-mov.pl.

This script uses the excellent free tool ffmpeg to slice and dice the input video with the appropriate cropping and produces a much smaller MOV file (i.e., QuickTime video) as its output.  You feed the script a CSV file indicating the start times, end times, and cropping areas of the various slices, and it does the rest.  There’s more information in the comment at the top of the script about how to use it.

Please comment here to let me know if you find this useful.  If the time I spent writing this script saves you some time or money, please consider sending a little something to my tip jar (but if you don’t, that’s OK too).

What’s amusing is that the way I found it through a Google AdSense ad that appeared alongside my complaint .

AVS Video Converter will convert from just about any format to just about any other format, including allowing segments of the video to be clipped during the conversion. Furthermore, it gives you complete control over all the encoding parameters in an obvious way, so you can play with them to figure out the settings that are best for you.

The same company makes a different product called AVS Video ReMaker, which does exactly what I wanted to do, i.e., allowing videos to be clipped without re-encoding them, but unfortunately it doesn’t support WMV files, so I had to go with AVS Video Converter, which does have to re-encode but produces a good enough result.

For anyone else needs to do what I did, I will share the specific settings I used to re-encode the WMV file produced by GoToMeeting, so you can use them as a starting point in your efforts to find the right settings for you:

• Encode to WMV
• Windows Media Video 9 video codec
• Original frame size
• 3 frames per second (the 30fps in the video produced by GTM is quite excessive for most purposes, although you can certainly increase this if you find that your converted video is choppy)
• Bitrate 1200, quality level 50
• Click on the “Advanced…” video codec settings button, enable VBR (which stands for “variable bit rate”), and choose the “Quality” setting with a quality level of 50
• WMA audio codec
• Mono, 16 bit sample size, 22050 Hz sample rate, 16 kbps Bitrate

This produces a video which, admittedly, isn’t quite as crisp as the original produced by GoToMeeting, but considering that it’s more than 40% smaller, I’m willing to sacrifice a little crispness. Of course, if you’re not you can increase the quality level.

Note that the company that produces the software is currently having a sale where you can buy an unlimited, perpetual license for all of their programs for one computer (if you buy a different computer, you have to buy a new license) for only $59.

I do have a few minor complaints about AVS Video Converter which I will mention here in the spirit of full disclosure:

1. When you edit the audio encoding parameters by hand and choose a combination that isn’t valid for the codec you’re using, it tells you that the combination is invalid but doesn’t tell you why or what to do about it.
2. It would be nice if the authors gave a bit more guidance about what all the knobs and dials mean, which might have saved me some trial-and-error time.
3. I found a few combinations of codecs and encoder settings which caused the program to crash or report internal errors.  I just worked around this by using a different codec or changing the settings, and so many of them are supported that it wasn’t a major obstacle.

Overall, though, it’s a pretty decent program, especially since it actually does what I need!