Best free hanzi to pinyin converter

[chinopinyin]

Which is the best free hanzi to pinyin converter?

1. Adsotrans http://www.popupchinese.com/tools/adso
   
2. ZDT http://zdt.sourceforge.net/
   
3. Google Translate http://translate.google.com/#zh-CN|zh-CN|
   
4. Dofufa http://chinesetopinyin.dofufa.com/
   
5. Chinese-tools http://www.chinese-tools.com/tools/pinyin.html
   
6. Other
   

[hanyu_xuesheng]

Just made a little test with these 4 sentences:

我受不了了。 Wǒ shòubuliǎo le.

发展经济，保障供给。 Fāzhǎn jīngjì, bǎozhàng gōngjǐ.

他个子虽小，力气却很大。 Tā gèzi suī xiǎo, lìqi què hěn dà.

他对滚石乐很感兴趣。 Tā duì gǔnshíyuè hěn gǎn xìngqù.

Wenlin (no "segment first"):

Wǒ shòubuliǎo 【◎Fix:◎le;◎liǎo;◎liāo;◎liào;◎liáo】.

Fāzhǎn jīngjì，bǎozhàng gōngjǐ.

Tā gèzi suī xiǎo，lìqi què hěn 【◎Fix:◎dà;◎dài】.

Tā duì gǔn 【◎Fix:◎shí;◎dàn】 【◎Fix:◎lè;◎yuè】 hěn gǎn xìngqù.

You have to fix 4 items

Adso http://www.popupchinese.com/tools/adso

wǒ shòu buliǎo le.

fāzhǎnjīngjì, bǎozhàng gōnggěi.

tā gèzi suī xiǎo, lìqi què hěn dà.

tā duì gǔnshílè hěn gǎnxìngqù.

2 mistakes: gōnggěi, gǔnshílè

Google http://translate.google.com/#zh-CN|zh-CN

wǒ shòu bù liǎo le。

fā zhǎn jīng jì， bǎo zhàng gōng jǐ。

tā gè zi suī xiǎo， lì qì què hěn dà。

tā duì gǔn shí lè hěn gǎn xìng qù。

1 mistake: gǔn shí lè; word boundaries not displayed

http://chinesetopinyin.dofufa.com/

wǒ shòu bù le le

fǎ zhǎn jīng jì bǎo zhàng gōng gěi

tā gè zǐ suī xiǎo lì qì què hěn dà

tā duì gǔn shí lè hěn gǎn xìng qù

5 mistakes: shòu bù le, gè zǐ, lì qì, gōng gěi, gǔn shí lè; word boundaries not displayed, no punctuation marks.

http://www.chinese-tools.com/tools/pinyin.html

wǒ shòu bùliǎo liǎo / le

fāzhǎn jīngjì bǎozhàng gōngjǐ

tā gèzi suī xiǎo lìqì què hěn dài / dà

tā duì gǔn shí / dàn lè / lé / yuè hěn gǎn xìngqù

You have to fix 4 items; 1 mistake: lìqì; no punctuation marks

http://www.purpleculture.net/Chinese-Character-Pinyin-Converter-Online.asp

wǒ shòu bù liǎo le

fā zhǎn jīng jì bǎo zhàng gōng jǐ

tā gè zi suī xiǎo lì qi què hěn dà

tā duì gǔn dàn lè hěn gǎn xìng qu

2 mistakes: gǔn dàn lè, xìng qu; word boundaries not displayed, no punctuation marks, C&P unhandy.

Conclusion: All tools are disappointing.

[chinopinyin]

hanyu_xuesheng, thanks a lot for the comparison but

Is the state of pinyin converters that bad or are you simply using somewhat tricky sentences to test them?

Does everybody think they are useless? or simply that they have problems, but it is still worth using them?

No converter is willing to speak up?

[chrix]

I'm happy with the converter that comes with the pinyin toolkit for anki. Of course there are some mistakes from time to time, but in a wordlist context is even less than in a sentence. One complaint I have with them is when you feed single characters into it, and the character happens to be a duoyinzi, it usually doesn't come up with the most common reading, but with the reading that comes first in its database. (ideally it would come up with all the readings, but that's not what it's been designed to do)

Chinopinyin, I don't think that the sentences used by hanyu_xuesheng were particularly tricky. I guess parsing technology just has a ways to go for Chinese.

[hanyu_xuesheng]

@chinopinyin

I chose of course some sentences that may have some "difficulties" for converters that are poorly programmed ;-)

我受不了了。 Wǒ shòubuliǎo le.

Two 了 with different Pinyin.

发展经济，保障供给。 Fāzhǎn jīngjì, bǎozhàng gōngjǐ.

给 is pronounced here jǐ, not gěi.

他个子虽小，力气却很大。 Tā gèzi suī xiǎo, lìqi què hěn dà.

个 in 个子 is pronounced gè, not ge.

他对滚石乐很感兴趣。 Tā duì gǔnshíyuè hěn gǎn xìngqù.

滚石乐 is rather seldom IMHO, 乐 must be yuè, not lè.

Even in simple texts you may find quite a lot of those problems. And especially for beginners the Pinyin should be correct!

[chinopinyin]

@hanyu_xuesheng

Thanks a lot for such detailed comparisons and explanations. Yours are the kind of posts well worth reading.

It seems clear that all current pinyin converters are flawed, though from your examples it appears that Adsotrans and Google have some comparative advantange.

But should I do use them with caution, being aware that they are not perfect, or simply do not use them at all?

[knadolny]

use them with caution

I use them from time to time but with extreme caution. There is nothing worse than studying the wrong way of saying something. I haven't found a good way for pinyin other than knowing the words before hand or verifying these results with another source like NCIKU or a native speaker.

For Google: wǒ shòu bù liǎo le

I'm not a fan of how they break up all the words instead of linking them together as they should be.

I will have to check out Adso.

Just made a little test with these 4 sentences:

Thanks. This is helpful.

[chrix]

knadolny, I think before relying on pinyin converters, learning the most basic duoyinzi/poyinzi is key.

[knadolny]

I think before relying on pinyin converters, learning the most basic duoyinzi/poyinzi is key.

I agree. I think you might have misinterpreted my post. That was basically my reply.

I use the converters when I need to pinyin-ize something "quickly". However, I usually spend a lot of time fixing all the mistakes. Therefore, it's not really that much quicker. This way, I just don't have to type in all the accent marks.

[chrix]

oh I was basically agreeing with you, so no misinterpretation there. I use converters for the same purpose...

[trevelyan]

Thanks for the correction hyxs. I won't speak for the others, but I think it's great that truly free alternatives like Adso are as good if not better than much more restrictive commercial software packages.

That said, I would encourage people to post about these issues on Popup Chinese or message me about them, as I wouldn't have run into this thread if a friend didn't forward it. The easiest way to fix any issues with Adso is to make edits directly in the Popup Chinese Chinese-English dictionary or email our tech team. I've just made the edits for these particular entries myself. It will take about a month for these to go live on the site, but they will fix the two issues pointed out for perpetuity.

Also a reminder that the engine can be downloaded here for those interested in doing more serious development work. Concerted efforts like this (at identifying weaknesses) are key to making long-term improvements to the algorithms used for word identification and spacing.

[chinopinyin]

I am also quite happy with Adso, although there is something that annoys me

If you copy a list of sentences and ask Adso to pinyinize them carriage returns are lost

So if I paste

请问，还有没有位子？

有，有，有。那张桌子没有人。

二位要吃点儿什么？

老王你点菜吧。

I get

qǐngwèn, hái yǒu méiyǒu wèizi? yǒu, yǒu, yǒu. nà zhāng zhuōzi méi yǒu rén. èr wèi yào chī diǎnr shénme? Lǎo Wáng nǐ diǎncài ba.

so I have to manually introduce carriage returns

This is probably something that could easily be fixed by the Adso Team.

Keep up with the good work!

[knadolny]

@trevelyan

I am about to start my next book project and after reading this thread am planning on using ADSO for "pinyin"izing the Chinese text. I'm not a computer programmer so I can't help with that, but maybe I can help you out some other way. I will be using ADSO to get pinyin for a Chinese story, Ah Q, and then will read it line by line, character by character, to root out all the mistakes.

I went through a similar process for my first book (see signature) and I can tell you this process was my least favorite part by far. I might have been better off without the converter. If ADSO can help make this process less painful I will be very grateful. Would you like to see my correction notes when I'm finished? My notes tend to be very haphazard, but might be worth something. I don't have date set, but will probably be done in the upcoming months.

more restrictive commercial software packages.

What are the commercial software packages? I am thinking of NJStar although I am not sure if it can "pinyin"ize characters since I don't use it. Are there others?

[westmeadboy]

I would like to do Hanzi-pinyin conversion programmatically (preferably calling some online service but also interested in algorithms I can use for offline) so what are the best options for that?

[Friday]

Is there any software or Web-based tool that can convert characters to pinyin (with tones) that can do it without errors?

[jbradfor]

without errors

Probably not, due to the large number of heteronyms. Would you settle for a few errors?

[Chris333]

Hey gang,

Haven't posted here for a while...been busy. Getting back into studying on my own now that I've stopped my tutoring and had a question;

This website came up in a discussion somewhere on this forum a while back, but I can't find the discussion now; http://xh.5156edu.com/conversion.html My question is this; Is there a website for converting the other way, i.e.; pinyin to hanzi? I've been emailing people in Chinese using Google translate and MDBG, but the copy/paste process is very tedious. It would be great if I could just type pinyin with tones and get hanzi output. I know what to say in spoken Chinese, 但是我看不懂汉字! :-)

Thanks!

[chinopinyin]

You can use Google pinyin input method http://www.google.com/ime/pinyin/index.html

This guide will help you set it up http://scattered-notes.blogspot.com/2007/04/translation-of-google-chinese-input.html

[Chris333]

Know of a good one for the Mac?

[hanyu_xuesheng]

The best input method for Mac is here: http://www.macqim.com/RegQIME.html

see also here: http://www.yale.edu/chinesemac/pages/input_methods.html#other