Pe̍h-ōe-jī transliterator

I made a (yet another?) transliterator for pe̍h-ōe-jī(Church romanization).
It’s a web app, and you can use it from here ->
I would be grateful if you could give me some feedback for improvements.

Thank you.


1 Like

Very cool. I love online converters like this. Thanks!

pretty good. I think you are only a few modifications away from turning it also into a Tailo transliterator.

there’s this tool online that came before yours:

I do think it’s interface could use some improvements, and maybe that’s where your tool could differentiate.

Thanks for the comments.

AAMOF, I’m not familiar with Tailo.
When I made this program, I deliberately avoid using character sequences ‘oo’ and ‘nn’ to
represent ‘ơ’ and ‘ⁿ’, so that I could use this program as Tailo transliterator. But, on second
thought, this program apparently misplaces tone marks when vowel is ‘oo’ in Tailo.

Maybe another program for Tailo is the right solution.
Is there any web sites (preferably in English) describing Tailo system in detail?

I’ve always used the wikipedia page: … ion_System

Thank you.
I incorporated licit syllable sequences which are described
in … ion_System,
so it should now have the ability to process both pe̍h-ōe-jī and Tâi-lô.

Did you write yours in javascript?

by the way, there’s a open source python script that converts POJ to TL and back:

Yes, mine is written in JavaScript, because I would like to use it with my
smartphone’s local environment.

Thanks for telling me POJ_TL, I’ll check it out.

Is this POJ?

From here

Yes. It’s POJ.

There were 8 tones, but in Taigi there are technically just 7.

Tone 1 2 3 4 5 - 7 8
Diacritics a á à ah â á ā a̍h
Value 55 51 21 32 24 51 33 54

Historically, how were 2 and 6 different?

Thank you.

How did you convert the Traditional Characters or the Pinyin into POJ? Did you do it manually? I clicked the links to the online translators mentioned above but I could figured out how to generate the POJ romanization the way you did

1234, and 5678 both matches to Middle Chinese 4 tones, pîng (平) sióng (上) khì (去) ji̍p (入). The tone names were chosen because these characters are representative of each of their assigned tones.

In Late Middle Chinese, each of these 4 tones split into a Im (ying) and a Iông (Yang) set, creating 8 tones.

Taigi’s 1234 tones are called Im-pîng, Im-sióng, Im-khì, Im-ji̍p.
Taigi’s 5678 tones are called Iông-pîng, Iông-sióng, Iông-khì, Iông-ji̍p.

Tone name
Im (flat) 1 2 3 4
Iâng (harsh) 5 6 7 8


Before Holo speaking people migrated to Taiwan, the 3 biggest branches of Holo are Tsuân-tsiu accent (泉州腔), Tsiang-tsiu accent (漳州腔) and the first mixture of the two, Ē-mn̂g accent (廈門腔).

Taigi in Taiwan are basically different mixture of the 3. Some region leans more towards the Tsuân-tsiu accent, some leans more towards the Tsiang-tsiu accent. In Lo̍k-káng (鹿港), there’s an old accent that’s as close to Amoy accent as you get, and a new accent that’s a bit closer to the rest of Taigi. All these accents have slightly different distribution of tones, and none of them have 8 distinguishable tones. Most have 7, some have 6.

I think originally Im and Iâng should have similar qualities. So I have a guess of what the 8 tones sounded like originally, but there’s no way to know if it’s correct.

Yes. I wrote those in their literary reading.

The first link in this thread is more like an online IME, you are supposed to type in POJ using numbers as tones, and it outputs the right diacritics.

The second link I posted converts between POJ and TL.

There’s a new Taigi online dictionary that’s more similar to what you wanted to do.

There’s also a new pure Taigi forum

1 Like