News for hackers. Stuff that matters.

A comment on Nvidia Accelerates Real Time Speech to Text Transcription 3500x with Kaldi. Parent comment

Re:Nvidia Accelerates Real Time Speech to Text Transcription... by zaroth on Monday March 18, 2019 @03:45PM (#19424454)
Kaldi certainly is a research project. Hopefully in the same vein as Postgres and not so much a toy implementation.

Could you conceivably run DeepSpeech and Kaldi side-by-side on the same audio sample and gain any kind of accuracy benefit?

To start with, errors would have to be uncorrelated, and you would need confidence scores on each word, which I assume you get.

[ Parent | on HN ]
Re:Nvidia Accelerates Real Time Speech to Text Transcription... by antt on Monday March 18, 2019 @04:14PM (#19424748)
Deepspeech can have multiple models running at the same time so I obviously see a benefit from using another system. Kaldi however is much faster to run in my experience and deepspeech is rnn based which puts a limit on how much you can pararalize it.
[ Parent | on HN ]
Re:Nvidia Accelerates Real Time Speech to Text Transcription... by jjwiseman on Monday March 18, 2019 @05:33PM (#19425543)
There are ways to combine multiple independent speech recognition engines to improve accuracy, e.g. ROVER: http://my.fit.edu/~vkepuska/ece5527/sctk-2.3-rc1/doc/rover/r...
[ Parent | on HN ]
Re:Nvidia Accelerates Real Time Speech to Text Transcription... by StudentStuff on Monday March 18, 2019 @03:54PM (#19424559)
I haven't seen any setups like you describe, but if you want to build a framework for comparing confidence intervals that use case would be possible.
[ Parent | on HN ]