On Wed, Jul 25, 2012 at 3:38 AM, Jonathan Jones
<eyeonus@gmail.com> wrote:
On Wed, Jul 25, 2012 at 3:18 AM, la .lindar.
<lindarthebard@gmail.com> wrote:
Anyway.
Would be able to say you are at least 90% confident that the file meets the requirements as listed on this page?
No. =D
It's done as an MP3 and that clearly asks for a wav, which is going to be a loooooooooooot bigger.
You might be able to type-convert if you don't care about the quality.
I'll upload a WAV version anyway. Let me know if you want/need it.
Don't bother. If I'm going to have to edit it- especially as much as it looks like I'll have to from your later comments- then I may as well just use the mp3 file I've already downloaded. Quality isn't a huge thing- as long as it's good enough to hear clearly, that's all that really matters. We're not making a platinum selling single, after all.
Basically, they want a single audio file with ~1 second pauses between words, and each word must be in order. There's a few more things than that listed on the page, but those are the absolute necessity.
It's mostly in order, but I backtracked in some places, so make sure it's correct.
You'll have to do the spacing and crap like that yourself.
Also I said each word *at least* three times, so you're going to have to pick the best take of each.
Absolutely the entire thing will have to be edited to remove the other two to six takes you don't want.
Yeesh, that's a lot of work. I'll get on it tomorrow, but I have no idea how soon I'll have it done.
So, I'm doing this in two parts. First, I'm sorting through the whole thing and exporting what I feel to be the best example of the >3 for each word. This will make it possible to update the audio for the Anki decks, which use a single mp3 for each entry. Second, I'll be pasting them all back together with exactly 1 second intervals of silence between them to give to Memrise Ben for uploading there. This might turn out to be more work for me in the long run, but not by much.
In order to do the audio for the gismu places deck/garden, I need "lo broda", "lo se broda", "lo te broda", "lo ve broda", and "lo xe broda". I really just need lo se te ve xe, but it won't sound natural if you say them individually and then they get pasted on.
doi.camgusmis. Can you write a script that can automatically parse the files together to make all the needed audio for the places, or will it need to be done manually?
I said each word a couple of times so as to avoid having to rerecord things if I balked a take.
That being said, if there's a word with no good take, let me know so I can redo it.
I'll be sure to do that.
--
mu'o mi'e .aionys.
.i.e'ucai ko cmima lo pilno be denpa bu .i doi.luk. mi patfu do zo'o
(Come to the Dot Side! Luke, I am your father. :D )