The AJATT Method

Index » General discussion

Topic closed
Reply #326 - 2008 May 01, 1:16 pm
Mcjon01 Member
From: 大阪 Registered: 2007-04-09 Posts: 551

tokyostyle wrote:

killerducky wrote:

... but I thought the idea was to simulate listening to speech.

You forgot that there are words and grammar that are for writing only and not used in speech.  Also speech has a rhythm and tone to help separate the homophones.

Mcjon01 wrote:

Certainly, they can make things easier, but they couldn't possibly be vital to comprehending the language.

So what does this sentence mean?
Mcjon01がいっちゃった!

I guess someone forgot about homophones ...

(I actually think the kana->kanji is good a idea though.)

I didn't forget about homophones, I just don't think they're as big a deal as lots of people like to say they are.  The huge majority of cases could probably be understood through context.  The example you gave has no context, so it's essentially meaningless.  It's not like I said Kanji should be abolished or anything, or that they're totally useless, just that they aren't strictly necessary.  I still think they're totally awesome and that they make a whole lot of things about the written language easier, though.

Reply #327 - 2008 May 01, 4:22 pm
sutebun Member
From: Oregon Registered: 2007-06-29 Posts: 172

Using spaces and certain rules regarding particle placement (and other common things), all kana probably could be read with little effort.

Without all that it'd be a mess though. Imagine trying to read books or papers! Never ending strings of kana except for periods and commas. You can't really find quickly where words start and end.

Anyway, its' a moot point. The language's entire writing system isn't about to change anytime soon.

Reply #328 - 2008 May 01, 5:32 pm
QuackingShoe Member
From: USA Registered: 2008-04-19 Posts: 721

There are some niggling points I want to pick at, but they're all beside this overriding point:
I find it painfully naive and arrogant to tell the practitioners of another language that their language would be easy if they'd just 'do this'. You don't have the basis on which to make this claim, whereas they do. I believe it wise to respect that the choice of the entire Japanese-speaking population in the representation of their own language is a wise one, until such a point as you have the same level of experience with the language as they do. After that, you can make such a claim. You'll still be wrong, but at least you'll have a basis on which to be so.

In the same way, it is best to assume that the strangely flowing script and right-to-left writing of semitic scripts comes with reason and is a wise choice. It doesn't matter that it's not how you do it. These other languages are not strange, inefficient languages that could be just as easily or more easily written the way our language is written simply because our language is our language. They're ancient languages, every individual part of which has evolved in unison over sometimes millenia to make a very coherent whole, and you don't have a right to say what would be an easy change. It's incredibly elitist.

I don't mean to personally offend. I want to specify that I'm not calling anyone naive, or arrogant, or elitist, or anything else I might have said. Only that this ATTITUDE is. One is capable of holding an attitude without being any of the things the attitude possesses. It's simply an attitude that I see a lot of, and I don't like to. Listen; you are not the default. Your race is not the default. Your culture is not the default. Your language is not the default. Your gender is not the default. Your viewpoint is not the default. Your taste in icecream is not the default. As learners of another language and culture, I think everyone here is made aware of this to some level or another. But it's easy to forget bits and pieces of it. Please, try to always remember.

Advertising (register and sign in to hide this)
JapanesePod101 Sponsor
 
Reply #329 - 2008 May 01, 6:11 pm
sutebun Member
From: Oregon Registered: 2007-06-29 Posts: 172

I 思 that is a 良 idea.

でも、もちろん、漢字を使いたかったら日本語で書ける。

But 使ing 漢字 in 英 text is also kind of 楽. It's a puzzle.

Reply #330 - 2008 May 02, 7:51 am
nac_est Member
From: Italy Registered: 2006-12-12 Posts: 617 Website

sutebun wrote:

I 思 that is a 良 idea.

でも、もちろん、漢字を使いたかったら日本語で書ける。

But 使ing 漢字 in 英 text is also kind of 楽. It's a puzzle.

I 笑'd.

Reply #331 - 2008 May 05, 6:33 am
joesan Member
From: Kawasaki Registered: 2007-09-18 Posts: 20

nest0r wrote:

So, for those doing the kana-->kanji sentences, how many of those do you do a day? I'm thinking I might have to revise the 25-30/day I had originally intended. My goal of 10,000 different words learned in 12 months (in sentence context) seems far away indeed. smile

I've been doing it for about two months now and on average add 10 to 15 new sentences a day. I spend about 1~1.5 hours a day at it though I was away for the Golden Week holiday and have a massive backload to work through now.

I used to write them out but that was taking too much time so now I only trace them out in my mind

Reply #332 - 2008 May 05, 3:53 pm
joesan Member
From: Kawasaki Registered: 2007-09-18 Posts: 20

nest0r wrote:

Do you do kanji-->kana/passive recognition as well, or just the kana-->kanji?

I review in both directions, though I find the kanji-->kana review usually takes very little time; I normally remember it straight away. I sometimes have trouble with whether an お sound is lengthened by a う or not which is why I keep reviewing in this direction as well.

Reply #333 - 2008 May 05, 7:25 pm
mr_hans_moleman Member
From: Toronto Registered: 2007-06-24 Posts: 179

This is why I started using audios from Japanesepod101.com. I figured that my sentences in my SRS should be reinforced with audio. That's why I started taking sentences from japanesepod101.com and then putting only the 'conversation part of the lessons' on my mp3. This makes so much sense to me, that I wonder why I didn't consider it in the past. I reply the audio over and over again. I call it, hitting the language at all sides kind of like the different punches of a boxer.

Reply #334 - 2008 May 05, 7:33 pm
mr_hans_moleman Member
From: Toronto Registered: 2007-06-24 Posts: 179

Oh yeah, I'm at my crazy period now with school being done and all. For the past few weeks, I've been doing reviewing at least 100 sentences per day (only kanji-kana). Everyday, I add about 50 sentences.

Reply #335 - 2008 May 06, 4:35 am
tomusan Member
From: Japan Registered: 2008-02-06 Posts: 79 Website

Not advocating piracy here, but if you are creative in your searches, its also possible to find the Misaki voice in Bittorent network

On the whole issue of switching to dictating. I got the impression that Khatzu decided to go that way for Chinese due to his dislike of the phonetic systems used to represent Chinese. Japanese doesn't have this problem, so although I can see the benefit I am not sure if it is strictly necessary.

Last edited by tomusan (2008 May 06, 6:23 am)

Reply #336 - 2008 May 06, 8:55 am
nac_est Member
From: Italy Registered: 2006-12-12 Posts: 617 Website

Not strictly necessary, but still a good exercise. I'm starting to notice a big gap between my (fair) reading-writing skills and my (poor) listening skills.
So doing dictations sounds like a good thing, at least for those who don't live in Japan.

Reply #337 - 2008 May 06, 10:17 am
tomusan Member
From: Japan Registered: 2008-02-06 Posts: 79 Website

I was also thinking it might be extremely hard to do at first, being able to write the characters just from listening straight off the bat is no mean feat.

Reply #338 - 2008 May 06, 2:52 pm
rich_f Member
From: north carolina Registered: 2007-07-12 Posts: 1708

I think it's a good idea, but not so much from a dictation standpoint as a pitch accent workshop sort of thing. When I studied Japanese in college, there was -0- discussion of accents. They just never bothered to mention it. Odd. I mean, there's a difference in the pitch accents of はし and 橋. (Something I had explained to me in-depth at Yamasa. big_smile) But not being in Japan really hurts in trying to figure out accents in words.

Out of curiosity, I sent an inquiry to Neospeech, and Misaki costs $350US. Ouch. Miyu isn't as good as Misaki, but that voice is only $35US. "Show" isn't bad either. He's the $35 male voice. The rep at Neospeech said that they didn't have a timetable for releasing Misaki to the public. The $350 version is the basic corporate version. (I'm assuming that's what Khatz did, and why he said he spent so much.) I looked at some other companies, and nobody is even close to Neospeech for quality. (Although some of the other companies' voices are so stiff and robot-y, they'd be great for mixing into music. big_smile )

Either way, implementation (as always) is the tricky bit. If you put a sound file for every single sentence... yikes. Reviews would take a very long time, I'd think. Then again, it would give you something to check your pronunciation against. (If you don't have any native speakers to work with, anyway.)

I'm not so sure I'd go the dictation route just yet, though. I might come back to that in a few months or so. Either way, it's a fascinating approach. Gotta give the man credit for trying new things.

Reply #339 - 2008 May 06, 4:29 pm
Ryuujin27 Member
Registered: 2006-12-14 Posts: 824

I was just checking out the Neospeech site and I was trying that 200 character sample on the main page. I typed はしと橋は同じ意味じゃない and tried both Misaki and Show.

To me, Misaki didn't sound that great, and was a little difficult to hear the difference between the two hashi's. Also, Misaki pronounced じゃない very, very weird. I actually had to listen twice to pick up what the hell she said.

Show, on the other hand, clearly demonstrated the difference. Plus, じゃない, spoken by Show, sounded quite natural.

So considering Show is only $35, I'd go for Show.

(However, I have not tested them in depth, so don't go by what I say)

*EDIT*

Ok, the difference is more noticeable when you trying to say ~じゃないよ

However, I still think Show does it better. Plus, for guys it will be better listening to a guy speak the lines.

Last edited by Ryuujin27 (2008 May 06, 4:35 pm)

Reply #340 - 2008 May 06, 7:17 pm
tomusan Member
From: Japan Registered: 2008-02-06 Posts: 79 Website

More customziable in that you can alter then speed and pitch of the voice to your liking from within text aloud.

Reply #341 - 2008 May 07, 3:31 am
ghinzdra Member
From: japan Registered: 2008-01-07 Posts: 499

About this text to speech device I was so enthrilled by the outlook that I made research on this right on the spot .


THE EXPERIMENT
And Is there a  best way to know how faithful to reality it is than asking to  the japanese themselves ?  so without any warning I made listen  news to several japanese teachers and collected their opinion.
There were 4 teachers  and 3 extracts : first one was Misaki reading a script of Fuji news Network  , secound one the very newcaster of  Fuji News Network  , third one Yomira podcast .

RESULT
*everybody agreed on the enjoyable voice of Misaki .
*Nobody agreed on the speed :
the 1st one thought 0 was really too fast  -7 was a radio speed , -6-4 a tv news speed , -4-2 normal speed
the 2nd one wasn't bothered at all  by the speed  I put it at 0
3rd one frowned from the beginning . She sensed something unnatural 
4th thought it was a bit quick .  -4 for her was a natural speed .
* the hatsuon was perfect for everyone except the third .
* when it comes to a direct comparaison
the 1st one prefered Yomira over the other one but she still prefered Misaki over the real newscaster of Fuji News Network !
the secund one was bothered by the lack of space but she thought it was due to needs of TV news . So she singled out Misaki .
Only the third one really sensed Misaki as unnatural . But maybe she figured out what I was trying to do as she's one of the teachers I asked to record the sentences for the KO2001 audio project . In the other hand  she seems very sensitive to this particular problem as on the very beginning of the project she offered me to record also the vocabulary besides the sentence  stating that the pronunciation really changes . After an explanation , she even pointed out part of the sentence where she thought it was unnatural .But the hatsuon of each word was normal according to her . She seems really reliable on this issue .
the fourth one  was also bothered the lack of space but until I told her she rejected the idea of Misaki as an artificial voice .



CONCLUSION
So it seems like Misaki while not being perfect is really something reliable .
. According to the people  it 'll range from unnatural to normal but quick .
(Once again ....smile) I just have to agree with Khatzu : you should look for a mix of natural voice and Misaki .
In fact  if you're able to catch up with her 0 speed when you're listening and if you're able  to speak like her  you'll be fluent for most of the japanese .
As far as I'm concerned I planned to mix yomira podcast which is easy to download and to follow  and Misaki  reading the news of fuji news network (after having used a lot of software  : goldwave ,getasfstream,etc... I eventually found how to get the streaming video file quite easily but it's still longer and more troublesome than downloading the yomira podcast.... the strong point of FNN  is that the script is totally faithful) .

Last edited by ghinzdra (2008 May 07, 3:37 am)

Reply #342 - 2008 May 07, 6:06 am
Sequa Member
From: Germany Registered: 2006-11-02 Posts: 40

Sorry this is a little off-topic:
Neospeech voices don't work on my computer (Vista 64bit) but on another computer (WinXP) everything works fine. I'm considering isntalling XP only to get it to work but maybe Vista isn't the problem but x64. Has anyone tried using Neospeech voices with Vista 32bit?

Reply #343 - 2008 May 07, 8:23 am
tomusan Member
From: Japan Registered: 2008-02-06 Posts: 79 Website

32Bit Vista works fine for me with Neospeech voices.

Back on the whole Kanji - Kana, Kana - Kanji discussion. I am considering just suspending all my Kana - Kanji cards for now, until I can at least read them well, them introduce them later (not too much later, 3 weeks or so after I add them). If I do suspend them I think I'd switch back to the original AJATT method, of copying out some of the sentences after I have answered them. Once I have down that a few times hopefully the kana-kanji won't take so long.

Perhaps its just me being impatient, and admittedly I have only just just finished RTK and have only just started with the whole sentences bit, but reviews from Kana - Kanji take SO long at the moment, my daily progress is pathetic.

Last edited by tomusan (2008 May 07, 9:45 am)

Reply #344 - 2008 May 07, 9:27 am
rich_f Member
From: north carolina Registered: 2007-07-12 Posts: 1708

@Ryuujin
Yeah, I had similar problems getting Miyu and Show to say ~じゃないよ properly, too. They both say ~じゃない perfectly, but the よ really throws off their cadence. Not sure what to do about that.

@ghinzdra
As I said in the AJATT sentences group-- awesome job on the research. Heh, if you were inclined, you could probably turn that sort of thing into a graduate research project. big_smile

I think a balance of native speakers and the TextAloud/NeoSpeech combo works pretty well. I use the TTS software to sound out sentences I'm reviewing just for kicks, or to work on my pronunciation... IF it sounds natural. If it doesn't, I'll take it with a grain of salt.

And as I mentioned on the AJATT gmail group, the program does have issues with certain readings, like in instances when 空 is alone by itself and is to be read から and not そら. When you change it to hiragana, the pronunciation is off. But in the long run, these are minor issues.

Oh, has anyone figured out how to make the dictionary editor work? In this thread:
http://www.nextup.com/phpBB2/viewtopic.php?t=3840
there was an interesting discussion of the pronunciations that the Japanese voices use, and how to change them.
There was an English translation of the interface for each voice's pronunciation editor here:
http://www.qvidnon.ca/TextAloud-Japanese-Editor.png
But I'm not too sure exactly how it works.

Reply #345 - 2008 May 07, 5:22 pm
Ryuujin27 Member
Registered: 2006-12-14 Posts: 824

Oh, and for those who want to test out Misaki or don't have $350 to spend at the moment... well, I'd just like to say that at some site somewhere you MAY be able to find SOMETHING about Misaki for downloading.

Something about torrents or something? Something like a small kind of nova?

wink

Reply #346 - 2008 May 07, 5:48 pm
Mcjon01 Member
From: 大阪 Registered: 2007-04-09 Posts: 551

Ryuujin27 wrote:

Oh, and for those who want to test out Misaki or don't have $350 to spend at the moment... well, I'd just like to say that at some site somewhere you MAY be able to find SOMETHING about Misaki for downloading.

Yeah, some site like Google. tongue

No, seriously, I wasn't even looking for torrents.  I just did a search for "Neospeech Misaki" to see if I could scrounge up reviews or something like that, and all I could find was torrents, as far as the eye can see.  It's like you have to try hard not to find them.

Reply #347 - 2008 May 07, 9:29 pm
Ryuujin27 Member
Registered: 2006-12-14 Posts: 824

Haha, piracy is funny sometimes.

*edit*

Also, quick question about Misaki.

I have it installed and everything, but now every time I type something and have her say it, it is all broken up by word, and most definitely sounds like a computer.

Did I do something wrong?

Last edited by Ryuujin27 (2008 May 07, 9:46 pm)

Reply #348 - 2008 May 08, 2:46 am
nac_est Member
From: Italy Registered: 2006-12-12 Posts: 617 Website

Ryuujin27 wrote:

I have it installed and everything, but now every time I type something and have her say it, it is all broken up by word, and most definitely sounds like a computer.

Did I do something wrong?

Did you leave spaces between the words? Misaki doesn't like that smile

Reply #349 - 2008 May 08, 9:01 am
penne New member
From: Sweden Registered: 2008-03-11 Posts: 6

Has anyone gotten Misaki to work under Vista? Gives me an error when I try to play a sentence sad

Reply #350 - 2008 May 08, 10:13 am
Ryuujin27 Member
Registered: 2006-12-14 Posts: 824

Nope, I didn't put any spaces between the words.

However, I think I forgot to set up TextAloud, so that may be the problem. I'll let you know after I do that.

Topic closed