3

Recently I noticed that Google and Apple have really high quality speech-recognition services. I was wondering about the state-of-the-art methods and techniques they are/might be using to achieve such quality. I already know that Hidden Markov Models can be used for speech recognition, but I was wondering if it's the used technique now. Because as I know HMM are old.

It would be great if you can link to some important recent papers that could include/descripe the used new methods.

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
Jack Twain
  • 7,781
  • 14
  • 48
  • 74
  • This 2010 review gives 166 references. http://arxiv.org/abs/1001.2267 – Zen Feb 06 '14 at 16:51
  • Android's is based on deep learning methods. http://www.wired.com/wiredenterprise/2013/02/android-neural-network/ – jerad Feb 07 '14 at 00:06
  • I found this [short overview](http://recognize-speech.com/acoustic-model/knn/benchmarks-comparison-of-different-architectures), but I am not entirely sure how "state-of-the-art" it is nowadays – Mr Tsjolder Oct 31 '16 at 13:08
  • I also found [this collection](https://github.com/syhw/wer_are_we) in the meantime... (I am looking for them as well) – Mr Tsjolder Oct 31 '16 at 13:23

0 Answers0