Alphabet Soup

[sabaijai]

Alphabet soup

Do you live in Chiang Mai, Chiangmai or Chiengmai?

An interesting new computer programme may have the answer

http://www.bangkokpost.net/education/site2006/cvau0106.htm

[Oswulf]

Personally, I can't see why the Bangkok Post bothered to publicise this (fairly trivial) program. The resultant transliteration might help ensure that all road signs to a place are spelt the same way, but the transliteration is no use whatsoever in letting you know how to pronounce the word - it completely ignores vowel length and tone.

[meadish_sweetball]

The software at thai2english can convert most common Thai words into three different transcription styles with a fair degree of accuracy. If anything, that site would be more worthy of recognition.

[Richard W]

Alphabet soup

Do you live in Chiang Mai, Chiangmai or Chiengmai?

An interesting new computer programme may have the answer

http://www.bangkokpost.net/education/site2006/cvau0106.htm

It says

"In Thai," he says, "pronunciation is relatively easy to predict, because we have rigid spelling rules. But in English…there are so many exceptions."

That's a bit harsh. Mark Rosenfelder claims quite good performace for English spelling to pronunciation - see Zompist sound changer:

Using sounds to find spelling rules

I've also used sounds to model the spelling rules of English. Here the input file lists the spellings of several thousand English words, and the "sound changes" are rules for turning those spellings into a phonetic representation of how the words sound.

Most people think English spelling is hopeless; but in fact the rules predict the correct pronunciation of the word 60% of the time, and make only minor errors (e.g. insufficient vowel reduction) another 35% of the time.

[mike_l]

Personally, I can't see why the Bangkok Post bothered to publicise this (fairly trivial) program.

I don't think anyone who's tried to write one would describe writing a Thai transliteration program as "trivial". There's a lot of complexities that aren't perhaps immediately apparent, and not forgetting the words need to be spaced beforehand to even be able to add syllables in.

That's a bit harsh. Mark Rosenfelder claims quite good performace for English spelling to pronunciation - see Zompist sound changer:

The big advantage I think is that there's a lot of research already done and useful tools available for English in this area (the CMU pronuncing dictionary, for instance), whereas for Thai there's very little by comparison. It seems to me it'd be possible to adapt that CMU dictionary to give pretty reliable English-to-Thai transliteration, it wouldn't be a trivial task but it seems doable. There used to be a webpage on the Kasetsart University server that would do it, sadly it seems to have gone now.

Edited August 27, 2006 by mike_l

[sabaijai]

The advantage of the Chula programme is that it gives you a transliteration into RTGS, the official system, hence is perfect for article titles, etc, to be used in citations and bibliographies for foreigers writing papers in English or other languages using Thai-language sources. Any other transliteration system, for this purpose, is non-standard. See the website for examples.