Hi
Does anyone know if it's possible to get timestamp-start and timestamp-end values of the MRCP input tag. Here's an example NLSML response from MRCP v2 document
<input>
<input mode="speech" confidence="0.5"
timestamp-start="2000-04-03T0:00:00"
timestamp-end="2000-04-03T0:00:00.2">fried</input>
<input mode="speech" confidence="1.0"
timestamp-start="2000-04-03T0:00:00.25"
timestamp-end="2000-04-03T0:00:00.6">onions</input>
</input>
For an input like this, the gateway takes only the latter value into account and ignores the confidence, timestamp-start and timestamp-end fields altogether. Here's what the gateway logs indicate for this portion:
016485: Oct 20 08:43:29.565: result
016486: Oct 20 08:43:29.565: interpretation
016487: Oct 20 08:43:29.565: input
016488: Oct 20 08:43:29.565: input
016489: Oct 20 08:43:29.565: pcdata: fried
016490: Oct 20 08:43:29.565: input
016491: Oct 20 08:43:29.565: pcdata: onion
016492: Oct 20 08:43:29.565: instance
016493: Oct 20 08:43:29.565: pcdata: fried onionvxml_asr_done: sematic meaning=fried onion confidence=99 utterance=onion input confidence=100 inputmode=voice
Since the utterance is set to the second word, the call studio form element only gets onion from the gateway and no other fields.
So is there a way to get all fields? Thanks
