is the sound computer generated?
Greetings. a quick question to duolingo admin. Is the sound computer generated for the spanish lessons? If so, cannot be changed for the future to actual human voice?
I think it works like this: They've recorded samples for the possible sound combinations in a language. A word like "Greetings" could be formed like this (sorry if I missed something) "Gr, ree, (ee), eet, ti, (i) ing, ngs". The computer knows how to use these samples to form words and other samples to leave spaces between words. The problem with recording human voice is that you would have to record every sentence separately. You cannot record words and stick them together as the intonation might be different for every word. A computerized voice can neutralise the intonation and I'm pretty happy with the way it sounds.
I hope I gave the right explanation.
thanks Lenkvist. That is helpful. Generally the sounds are good but sometimes they are hard to hear (happens to me especially when I have no hearts left and I am two or three away from finishing the exercise haha ). More importantly, I would be spending more time on duolingo if the voice was a bit more human like. It becomes difficult hearing that sound for a long time.