The Microsoft 2016 Conversational Speech Recognition System · HackerPapers discussions

The Microsoft 2016 Conversational Speech Recognition System — mentioned by cbasoglu

Fri, 16 Sep 2016 05:44:01 GMT

From the linked arxiv paper, http://arxiv.org/abs/1609.03528 this is a very interesting use of CNTK to adapt image CNN techniques to speech recognition. Surprising that CNNs worked so well on speech audio. Full disclosure: I am a MSFT employee.

Paper and discussion history. A mention is not an endorsement.

The Microsoft 2016 Conversational Speech Recognition System — mentioned by rngesus

Thu, 15 Sep 2016 05:30:31 GMT

The paper itself can be found here https://arxiv.org/pdf/1609.03528v1.pdf quite interesting to see that the failure rate is lower than the average human failure rate, can't wait to see how this will improve over the coming years.

Paper and discussion history. A mention is not an endorsement.

The Microsoft 2016 Conversational Speech Recognition System — mentioned by nshm

Wed, 14 Sep 2016 23:52:01 GMT

https://github.com/kaldi-asr/kaldi/blob/master/egs/fisher_sw... In Microsoft paper http://arxiv.org/pdf/1609.03528v1.pdf Table 5 it's a line "Povey et al. [19] LSTM". http://www.isca-speech.org/archive/Interspeech_2016/pdfs/059... Interpolate it to RNNLM column and you'll get 7.8. See also http://www.isca-speech.org/archive/Interspeech_2016/pdfs/047...

Paper and discussion history. A mention is not an endorsement.

The Microsoft 2016 Conversational Speech Recognition System — mentioned by nl

Wed, 14 Sep 2016 22:42:37 GMT

It measures against what humans write down when they listen. Humans generally use context, so I'd expect that they would record "cloud" if that was appropriate in context. The Microsoft system also does this: it uses language modelling to attempt to model what word is more likely in a given context. This gives them a 18% word error rate reduction (see section 7 of the paper: http://arxiv.org/pdf/1609.03528v1.pdf).

Paper and discussion history. A mention is not an endorsement.