Speech recognition software - needs to be "trained"?

Hi,
I'm wondering if there's any speech recognition software that doesn't need to be trained by the person whose voice is being recognized. For example, I have recordings of lectures, not my own voice, that I'd like to be able to convert to text.

I wouldn't mind having to do a considerable amount of editing, if it's doable at all.

Thanks,
Karen

 

http://www.lumenvox.com/resources/tips/types-of-speech-recognition.aspx Hope this helps from what I read this should be what your looking for.