I just did a 1-hour calibration of my voice recorder (Olympus VN-5500PC), ready for some audio mapping. It created a file 3601.5 seconds long, so that's a calibration factor of 3601.5/3600.0 = ~1.00042. Given that I probably didn't press the buttons to better than 0.5 s accuracy, that factor could be anything from 1.00028 to 1.00056. Still, in any case, that's better than 0.06% error.
Umræða
Athugasemd eftir Rhubarb sett inn 25. ágúst 2009 kl. 23:44
That's a great idea, I should do this for my digital voice recorder - it'd make surveying by bicycle a lot safer than pulling out a camera all the time.
I don't think it'd be too hard to figure out how to do audio mapping in JOSM - this'll give me something to do this upcoming weekend :P
Athugasemd eftir JeffB sett inn 26. ágúst 2009 kl. 11:05
I would certainly recommend it. It made a huge difference to my note-taking ability, not to mention safety, and it only took me a few minutes to get the hang of the audio features in JOSM. It's impressive to watch the little orange arrow moving along the road in perfect time with the audio (thanks to David Earl's software, I think). And you can practise your radio voice for when you become famous :-)
One thing you need to be aware of with this approach is the vast amount of information that you will collect. My first 1-hour survey is taking me several hours to listen to and transfer to the map.