If you can stream the signal through a computer there may be a way to use ffmpeg to process it in nearly real time. I remember reading some guy's blog post a while back where, if I recall correctly, he was recording signals and then using ffmpeg to convert the audio only to a Dolby format that would play in full 5.1 audio through his receiver. I think he had set it up so that ffmpeg ran as a post-processing command in TVHeadEnd, so the entire show had to be recorded before the conversion could take place, but I have heard that ffmpeg can be used on real-time streams so I've always wondered if it would be possible to do something like that in real time. Don't remember where I saw that now, and it probably wouldn't work anyway in a receiver that can't run ffmpeg, but if the receiver is running Linux internally then it might be possible for the manufacturer or US distributor to add ffmpeg and a way to do this conversion, either on recordings or in real time, if they had any interest in doing such a thing.
If anyone is seriously interested in this and a Google search doesn't locate the page for you, I might be able to find it again, but it would probably take me a while. If I recall correctly it wasn't that complicated, he was just calling ffmpeg and passing the video and data streams unchanged, and mapping the audio to Dolby 5.1 (but only on recordings, not in real time). I also seem to recall that the audio layout was similar to that used on other feeds (like you sometimes see on the slates on the feeds channels) except divided up among three APIDS, so the three APIDS had the channels like this: (Left Front, Right Front) (Center, Low Frequency Emitter) (Left Surround, Right Surround).