Hi Arsen,
Thanks for your help on the DTMF mixed recognition.
We have a requirement to be able to stop a RECOGNIZE session part way through in a way that makes the server return a recognition of the voice data sent up to the point the stop happened. This is without having to wait for the "speech-complete-timeout" seconds before we get a recognition result back. Is it possible ?
Looking at the code we have the following which might be relevant:
RECOGNIZER_GET_RESULT
RECOGNIZER_STOP
mrcp_application_session_terminate
If we sent a RECOGNIZER_GET_RESULT would it just return a result of the recognition so far ?
Thanks for your help.
Ed James