How does Microsoft speech recognition API compare with Google Cloud Speech API in terms of speech recognition accuracy (e.g., in terms of word error rate, character error rate (CER), or sentence error rate (SER))?

I'm also interested in other online speech recognition APIs.

I have crossposted the question at:

  • 1
    This is clearly a "rapidly changing event", so any answer would become obsolete every week or so. :-) Commented Oct 28, 2017 at 0:17
  • 1
    @bytebuster we could timestamp the evaluations :) Commented Oct 28, 2017 at 0:20
  • 1
    What about those built into Chrome and Android? FWIW I bet the difference will be in speed, price, languages, and things like robustness to background noise and other unlanguage that a benchmark may not capture. Commented Oct 28, 2017 at 21:39
  • @A.M.Bittlingmayer I'm also interested in those, but to a lower extent as they cannot be used in any application, unlike online APIs or open source ASRs. Some other factors: speak accents, and ability to take into account the context. It's hard to benchmark indeed :) Commented Oct 28, 2017 at 22:30
  • @FranckDernoncourt I have one data pipeline that invokes Chrome instances programmatically for some other Chrome functionality. It has bad latency but good enough throughput, ie good for preproc and data gen but not viable for live requests. Commented Oct 29, 2017 at 7:18


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.