Looking for a consultant with a deep technical background in implementing speech recognition.
We have an existing app for learning foreign languages:
There are many different languages available.
We currently encourage users to speak a phrase and process and match the recording of the user to that of a recording of a native speaker speaking the same phrase doing analysis using PLP and MFCC.
We have found this to yield reasonable but not great results. We would like to hire a consultant to advise and help our engineers implement a better system that will more accurately ascertain whether user has correctly spoken the requested phrases.