h a l f b a k e r yThe leaning tower of Piezo
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Please log in.
Before you can vote, you need to register.
Please log in or create an account.
|
This idea is motivated by the fact that with mp3 at 16 kbps I could fit 100 hours of spoken audio onto a CD and have it sound quite good. 100 hours is enough for for about 35,000 words and phrases in the standard language practice tape format:
"Good day."
"Guten tag. ... Guten tag."
Getting
a bilingual speaker to translate and record that much audio would take a just few hundred hours.
Pimsleur language tapes, as an exmaple, cost $200 for just 15 hours of audio. For many languages (e.g. Mandarin?) there are no advanced level tapes available.
So, I propose that we produce mp3-based language practice readings where the focus is on *quantity*, not pedagogical quality. Many languages, speakers with many different accents, many, many words. As I see it, there are two possible approaches:
Closed, commercial: hire translators to translate the phrases and native speakers to read them. A technician with some proficiency in the languages cuts up the recordings into individual samples, which are stored in a big database. The samples can be mixed and matched automatically to produce mp3 CDs at any desired level, with any number of repetitions or different speakers. Sell these CDs for $100 each, possibly offering customers the ability to customize their CDs.
Open, non-profit: You start a website soliciting donations of translations and readings. Get your friends to help you fill in the first few tens of hours. You'll be in charge of checking and filing the clips when they come in. Produce various concatenations of the clips as long mp3s and distribute for free via bittorrent to many happy people.
Books on Tape on iPod
http://www.halfbake..._20Tape_20on_20iPod
Related to this idea. [pdg37, Oct 04 2004]
Pimsleur tapes
http://www.languagetapes.com/
Pimsleur language tapes [pdg37, Oct 04 2004]
<off topic>
http://www.halfbake...dio_20_22Gallery_22
for [bwv] please add sample of your voice - or attach a recording in an e-mail to me and I'll add it. [Worldgineer, Oct 04 2004]
(?) testing this idea
http://www.deconfus...org/tmp/smallclips/
Here's a really boring set of intro phrases to get us going on this. Please translate into your native language and post a link to the raw file. [pdg37, Oct 04 2004]
[link]
|
| |
PS if people think this is a good idea I'd be really excited to try doing it one way or another. The help this would give me with languages is well worth the time it would take. |
|
| |
From teaching experience I can only comment that having several speakers in one language can only slow down the process, as this kind of learning is from repetition, and a different accent, i.e.pronunciation of a word, will confuse the learner. One speaker per language ok, a different speaker for the various dialects, ok. |
|
| |
I like the technology, it keeps getting better doesn't it. |
|
| |
Also I like your suggestion of a separate translator. Much better quality there. |
|
| |
Am I misunderstanding about the various speakers? |
|
| |
I think I agree with you on a single speaker for the introductory class, but for becoming conversational I would like to be exposed to different accents. In Mandarin, there is a huge difference between local Taiwan and Beijing dialects, which each differ in their own way from the "standard". Also, hearing male and female voices seems important. Maybe you would suggest that the different voices be separated into different lessons? |
|
| |
[bwv61] what? When speaking Portuguese do you only use obscure phrases? |
|
| |
I agree that in advanced studies the variety would be an advantage. |
|
| |
Oh Boy [bwv61]!, we have *got* to get you on tape!. |
|
| |
If your going to pick, why not pick complimented? |
|
| |
Audio tape! Audio! - I'm attempting to construct a sentence that maximises the letters mentioned! |
|
| |
Perhaps we need an audio babelfish. |
|
| |
//could get you in a lot of trouble using this informal format// I have no idea what you're talking about. |
|
| |
Tip for travelers to Portugal: When stopped at airport security, this complement is sure to get you through without stopping: Eu penso que você é feio e cheiro de peixes velhos |
|
| |
We would definitely have English speakers for the English. Heck, I'll do it, I've always wanted to go into radio... One advantage of doing this on the computer is that we could record phrases in English, Japanese, and Russian, and easily produce all six possible lessons: E->R, R-E, J->E, E->J, R->J, J->R. |
|
| |
Isn't this just a new 'flavour' of language CDs and tapes, not a new idea? |
|
| |
(Off-topic, marked-for-expiry)
[pdg37], ni shr Taiwan-ren ma? You know about the accents, and you knew about the traffic lights too. |
|
| |
[spacemoggy] wo bu shi! But I'm flattered you thought so. I know about the traffic lights, too - but when did I mention them?? |
|
| |
Great, [bwv61], I guess I'm going to learn Portuguese now! I had better go plan some intro lessons and buy a mike. |
|
| |
I think it's a new idea...obvious and non-patentable, for sure, but is anyone doing it? I don't think so, and I think someone should do it. |
|
| |
Okay, for testing purposes, I have learned how to do sound recording on my little voice recorder, and linked up some example intro phrases. I'd love to learn to say them in Portuguese; if you could translate them and record a file, I'll start playing with the software. Record at 44.1khz, speak clearly, and let me know if any of the phrases are not translate-able or if you have new additions to make. Then send me the raw sound file. |
|
| |
Notice that ogg really outperforms mp3 here. For each, I adjusted the quality up until I heard no more annoying artifacts, and got a smaller ogg file by more than a factor of two! Anyway, even with mp3, at this rate I calculate we could fit 10,000 practice words and phrases on a CD. |
|
| |
Well, if I sound less robotic after the first three or four, the next ten thousand should be mostly okay. It's hard to speak clearly and still sound human! |
|
| |