<div dir="ltr"><div class="gmail_quote"><div>Bro Peter,</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">> Of course, the time and work it would take to put all of that into the<br>
> Chinese module is another issue...<br>
<br>
That indeed is the crux of the problem. Having said this - I guess that<br>
for most Chinese signs the transliterations are known and described -<br>
even if they are manifold - and that should in turn make the matter<br>
scriptable - more or less:<br>
<br>
You assign to each Chinese character the variety of transliterations -<br>
then run the equivalent pinyin text across it and use this to assign the<br>
pinyin to the characters. I would think the majority should become<br>
automatic.<br></blockquote><div> </div><div>Most of it should be an automatic process, but then there are variations that, while standard (and yes, most certainly known), require some pretty intelligent programming to distinguish which variation is demanded in a particular context. So initial process could be automated, and then a person could go through and check. This would still require verse by verse reading of the thing to make sure everything was right. The deal is that some characters (many, in fact) have multiple pronunciations, some of them simply a tone change, others completely different. One character might have one pronunciation that would tend to occur, say, 90% of the time, but that other remaining ten percent may not be correctly rendered in the automation process.</div>
<div><br></div><div>Matthew Patenaude</div></div></div>