Hello,
I believe there is a problem with streaming-recognition generate-output-audio parameter with combination with solution from this post https://groups.google.com/g/unimrcp/c/3rxSUkt5hFw. In Dialogflow ES there was output_audio_uri value which was sent along GDF Plugin response to the application server, it contained an url to the UniMRCP saved generated audio from Dialogflow. Now this option is not being sent along. I did not have time to try it out with tag format xml/semantic, but still I would expect to see something send along the response, regardless of format-tag.
What I was able to see was a MRCP message with uri-waveform header with the correct url to the DF generated audio. The VXML documentation says it would be possible to use it after setting <property name="recordutterance" value="true"/> and access the url via application.lastresult$.recording, but what you get on our platform is the actual audio payload. When this is parsed further to submit, it crushes the application.
Is there some other workaround for this, or am I missing something? I checked the GDF Plugin wiki, but there are no new changes about the behavior of generate-output-audio mentioned.
Kind Regards
Maciej Krzyszton
atip GmbH
--
You received this message because you are subscribed to the Google Groups "UniMRCP" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unimrcp+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/unimrcp/665fc518-fc2a-4710-9a03-b9aef5870e65n%40googlegroups.com.