http://www.skritter.com/forum/topic?id=69162157&comments=5 - I've already made some proposal about sentence system

Here is another one that might be useful as well: it is very very useful that the characters and words are pronounced. How about making the example sentences soundable with a button click? This way we would learn to speak the whole phrases. Those like 世界上有七大洲 might be very useful.

What would you tell guys?

We've discussed the idea of doing text-to-speech to generate audio for some of the sentences, but held off because we're ambivalent about the value of text-to-speech for the learner. Some older threads on it:

http://www.skritter.com/forum/topic?id=48367530
http://www.skritter.com/forum/topic?id=56809854

It would be more valuable to record sentences, but that would be a lot of recording to do!

Yes ! A short phrases like the one you have cited would be very helpful.
At least for the like of me who has bad ear which has been trained to
Korean and Japanese pronunciations of Chinese characters.
Kimyong.

Maybe a link to Google? The text to speech is not bad at the level of short phrases.

http://translate.google.com/translate_tts?q=%E6%88%91%E7%88%B1%E4%BD%A0%EF%BC%8C%E7%88%B1%E7%9D%80%E4%BD%A0%E5%B0%B1%E5%83%8F%E8%80%81%E9%BC%A0%E7%88%B1%E5%A4%A7%E7%B1%B3

They also have pinyin, click "Read Phonetically"

http://translate.google.com/?hl=en#zh-CN|zh-CN|%E6%AF%8F%E6%AC%A1%E5%88%B0%E4%BA%86%E5%A4%9C%E6%B7%B1%E4%BA%BA%E9%9D%99%E7%9A%84%E6%97%B6%E5%80%99%E6%88%91%E6%80%BB%E6%98%AF%E7%9D%A1%E4%B8%8D%E7%9D%80

I don't imagine I'd use this all that much personally, but I suppose some might.

You need to copy and paste those links for them to work.

Sure, recording sentences is the best thing. And that's a hell lot of work. But c'mon guys:

* you don't need every single character, but rather those that are shown to the beginners - say Pimsleur, or first levels of HSK

* a single sentence has several characters covered

* you have the whole community under your fingerprints :)

I think if you open up example sentences together with a recording feature or upload ability, this should not be too much effort.
Some 1on1 classes here in China do this for their students, so there should be lot of native speakers available as long as they give permission.
This might be the only big issue that people could upload unauthorized stuff from existing books or schools who do not want that used. Legally you should be able to stay clear of that though as sites like Youtube etc. are able to operate too.

If you allow language schools to link to their school for every recording they provide (somewhere in the pop-up) you will have the sentences filled by native speakers in less than no time.

Quality control could be via you or the community rating the recordings.

Well, in fact the contribution model might be simpler yet effective. For instance:

* Audio submissions are allowed only to registered users that have reached a particular high level (both in writing and in tones especially)

* Some rating system should be provided. Yet, not everyone can give the same rating (depending on a level).

* Reward system (for recordings only) should be smart. Say, some amount of reward points for 100 or 200 (or more) sentences. Single sentence submitters might be welcomed, but to gain points - might need to work harder :)

The added value would be that some sentences will be even pronounced by more then a single radio person. Thus, our ears will be trained much better. We can't come to China right now, but you guys can bring China to us :)

PS: those are nothing but long-term ideas, I'm sure you've got a plenty of more important features to work on :)

*cough*kanainput*cough*