Hi, don't you think it would be great to have a voice incubator?
We all know that robotic voice is a must-to-be-changed on Duolingo. In order to send an application a contributor should read a text (it could be shown on the screen when you submit your application), a contributor should have a good microphone and free time) I think that it is something that may solve the problem with a robotic voice.
Who likes the idea please upvote it so that people from Duolingo could notice it.
1 a long while ago they actually did try this but they got bad results (is what ive heard) which doesn't mean they won't in the future but they have tried this before
2 even if they do find someone with a good microphone you would have to have someone with plenty of time because they would have to have the same person do vocals for every single sentence (there's a lot of them)
3 with a human voice it would be to varied if they were sick one day or there voice wasn't right or a large number of things it would cause to much variation and may confuse people on how to really pronounce certain words
4 you would have to have mods to review every single thing recorded to make positive it was at good standard of quality
my advice is if you really want a real speaker either use skype or better yet find a native speaker or if you just want a better vocal software there are many free ones on the web wear you can type in your sentence and if will speak it
its a good idea but at this current time it would be to time consuming and strenuous on the small amount of crew
My take on these points:
I've read elsewhere that this is actually on the to-do list. Maybe the problem in the past had to do with how so much is in flux here?
I think it would be beneficial to learners to have multiple native speakers on the course, anyways. That's the experience I've had on Memrise, but there it's just words, not full sentences.
I don't think this is much of a concern. Don't record if you can't speak well. You would have to look, though, at things like dialect (since this site teaches American dialects of English/Spanish/Portuguese, for instance) and also speed/enunciation.
If people were cleared first to join the team (and thus became moderators themselves) I think we could mostly circumnavigate this issue, although there would still need to be some level of quality control, yes.
This is a great idea but as blohrding points out it is, for many reasons, a difficult one. Luis has said previously in his AMA a few months ago that this is already on the to-do list. It's worth remembering that the courses have many thousands of sentences that would take probably hundreds of hours to record. However we can still be hopeful for the future :)
I think it's a good idea. In the 1980's when I was working in a Chinese restaurant, a cook there, who was from China, gave me some books and blank cassette tapes and asked me to read the books for him onto the blank tapes. I met him again many years later and was so surprised to hear him speaking English just like someone born and raised in Canada!
On the OLD Livemocha website, (not the website they have now) people were able to speak in their review of the person's English, for people who were learning English. In general, the learners really liked it when someone left them an audio comment and read it for them so they could hear native speakers. There was always a variety of English accents: British, Australian, American, Canadian, ....etc....
In one case, an American woman read something for a Chinese girl and the girl replied, "Which way should I pronounce "a"?" because the American woman was pronouncing it like "ay"? I wrote to the Chinese girl and told her not to worry about it, saying, "That's how Americans talk." and she wrote back to thank me.
Actually I love the idea. There could be one team for building the course, one for building the different voices (in one High language - mixing up American and Australien English would not that be great) and everything for free - that would be enchanting.
I think the worst part is microphoning everything with high-quality microphones - very few persons have such, This is probably much harder than implementing all the stuff into DuoLingo.