Seq2Seq on speech recognition

farizrahman4u / seq2seq

Sequence to Sequence Learning with Keras

GNU General Public License v2.0

3.17k stars 846 forks source link

Seq2Seq on speech recognition #138

Open ajason6208 opened 7 years ago

ajason6208 commented 7 years ago

I am a newbie on attention based model. I want to make an phonetic recognition in your library, will it work in practice ? Suppose I have 10000 training data, consist different size of frames, each frames is 39 dimensional.

Could you give me some advices for implement these system ?

sculyi commented 7 years ago

@ajason6208 Hi, I am now trying to use this lib to finish ASR, could you give me some advice?