Haven't used skritter for many months and just wondered what changes had occurred for Japanese.

I love Skritter but stopped using it because I was waiting for example sentences and the inclusion of pitch changes. Regarding the latter, if re-recording all the incorrect audio was impossible then a visual indication of what the pitch movements in the word are would have done.

I don't know if any progress has been made but I couldn't see any obvious changes when I checked just now.

Just in case the original point has been forgotten, I'd like to remake it - pitch needs incorporating because what skritter says is often incorrect and a piece of vocab is incomplete without it, (just as an individual character is incomplete without its tone in mandarin). There are only a finite number of patterns so you could just tag each word with the correct one (or let us do it as we go and give us credit against our bill).

Words are not homophones in Japanese just because their kanji share the same hiragana - that is something that all textbooks are guilty of promulgating except for the 文化日本語 series, (and even they only give the issue minor billing through inflections marked on words in the vocab book).

It's a real feature of the language that impedes communication if it isn't known, and it needs dealing with at the same time and at the same level of importance as the vocab itself.

To make an analogy in English (using stress rather than pitch), the difference is a bit like hearing: 'could you help me with some light housework?' compared to 'could you help me with some lighthouse work?' One error might be no problem, but a sentence full of them will be difficult to cope with for even the most charitably-minded conversation partner.

James

Those changes haven't been implemented yet. But as Chinese is a tonal language I am sure the people at Skritter have thought about it for Japanese. Rather than thinking about the tones for individual words I would recomend listening to complete sentences and repeating them as the tones for individual words will change depending on how they are used.

There are some pitch rules which govern how pitch falls across a sentence as a whole, or when particles take the pitch of the word they follow etc. which can be quickly learned by imitating people, as you say. But you can't learn every word's individual pitch accent that way. It's an inconvenience but you just have to learn them when you learn the word if you want to go far beyond intermediate level conversation.

All this is well-documented in e.g. dictionaries for TV presenters that have the NHK-standard pitch patterns in, for example.

By going beyond intermediate level conversation do you mean sounding like a native speaker? I have viewed conversation level to be a combination of known vocabulary and the ability to use that vocabulary. In college I studied Japanese the Spoken Language book 1. It does exactly what you want. However there is a reason people stopped using the Audio-lingual method that the book uses. And that reason is it's boring as hell. I would recomend studying using the Audio-lingual method for anyone wanting good pronunciation.

James, those features are still on the list of things we want to do. It's going to take us more months before we can do them, though. The plan for audio is just to record word-specific rather than just kana-specific pronunciations for all of the words we have. The plan for example sentences is to build this awesome example sentence system with most of the features that people have been hurting for, totally replacing the current sentence system.

Ddapore: Well, there is quite a gap between intermediate and native and I don't care at all about getting to the native side, but I want to do a bit more to make a complex conversation less of a struggle for all concerned.

Nick: thanks for the update. Both things sound good.

Regarding example sentences (a bit off topic, I realize), the link to Tatoeba has been very helpful to me in understanding word usage.

Hi everyone. I have been using Skritter as a complementary learning tool for the last few months and I'm very pleased with it. Nevertheless, the issue James Sharp brought up seems to be important enough to adversely affect the quality of the learning material, which worries me a lot.

As a native Spanish speaker, I understand how the intonation of certain syllables helps differentiate words. Is this the case in Japanese? If so, I think this is a major issue, of which all Japanese-Skritter users should be aware ASAP.

I look forward to the mentioned word-specific recordings, but I understand you guys have other things to do right now (we all want that beautiful, beautiful Iphone app).

Thank you in advace.