FYI, updated version of my slides in this presentation is also available at my Research@Google page. research.google.com/pubs/pub45882.html
@ananthakrishnank3208
29 күн бұрын
6:10 What does P(lambda) even mean? "Probability of a model", does not make sense.
@GANESHMCS51
4 жыл бұрын
concatenative shows jumps,. but wavenet shows so much distortion. any reasons for this ?? HMM and LSTM looks much better. One cannot really agree that the concaenative ones shown is production quality.
@sravanthinuthula7032
6 жыл бұрын
i am a beginner in machine learnig. i want to implement text to speech synthesizer ...so could you please tell in which areas i need to be perfect in...can you give a road map in order to implement TTS(like LSTM , RNN ,G2P,segmentation model)
@Drodothusfaram
5 жыл бұрын
hello, im beginner too, did you get one?
@romeozegna6543
6 жыл бұрын
My favorite text to speech product is Neospeech. Their technology is so natural sounding!
@InfinityDz
6 жыл бұрын
The biggest difficulty resides in the fact that almost all the tools are in Linux
@Fuckutube547465
6 жыл бұрын
How is that a difficulty?
@InfinityDz
6 жыл бұрын
I'm still trying to figure out how to install all the tools properly. I just installed a Virtual Machine today, things seem to work a lot smoother than with cygwin
@tomservo5007
6 жыл бұрын
infinityDz, it's worth to learn linux. I use it every day at my job and at home. Thank god that I can. It makes my life so much more easier. Good luck.
Пікірлер: 10