Is automatic video captioning possible?

There are some manual captioning services around, captvids.com is one, that do manual captioning of videos. That’s cool but I’m wondering if it’s possible to do automatic captioning?

There’s a lot of speech recognition software around, Dragon Naturally Speaking for example, that do an amazing job. It’s probably 7 or 8 years since I’ve used a voice recognition system and even back then they were near perfect – they can only have gotten better since then.

The thing with speech recognition software that wouldn’t lend itself to captioning is that you have to train them. You are asked to say certain words and sentences and the software records your voice and learns what you sound like. It then uses what it learns to make such a good job of turning voice into text.

Does anyone know if there’s anything out there (website or PC software) that captions videos automatically for you?

I’m looking into the possibility of trying to write something to do this and I don’t want to bother if it’s already been done. I think it would be possible to write something that listens to a video playing through once and produces some text. That text is not going to be perfect because there has been no learning but it might be good enough to produce captions that are usable.