[Olympus developers 180]: Re: problem with apollo

[Antoine Raux]

Hi Thomas,

I haven't looked at Olympus in more than a year but here is my guess...

This problem comes from the fact that Apollo is trying to do endpointing 
based on partial ASR hypotheses and (most probably) not getting any. The 
quick fix is to change the endpointing method in Apollo.cfg so that it 
only relies on energy rather than advanced method (like FSTTM). The 
potentially better fix is to figure out why Apollo is not getting 
partial hyps, which could be due to AudioServer, or to hub configuration.

The last solution is to learn to be very concise when you speak to the 
system.

Hope this helps...
antoine

Thomas Harris wrote:
> I'm having an endpointing issue with the olympus trunk on an 
> application built with mostly meetingline configs. Everything is 
> getting prematurely endpointed. It looks like a bug in Audioserver or 
> its configuration. Has anyone seen this, or better yet (not seen it 
> with some particular version and config).
>
> From Pocketsphinx:
> [STD at 15:55:47.691 (10909)] Processing command: engine_begin_utt 000 6070
>
> From Apollo, 40ms later:
> [WAR at 15:55:47.731 (10951)] Too much time since last partial result 
> arrived (4381>2000),endpointing here, regardless of VAD.
>
> I'm finding it hard to say much in 40ms :)
>
> Thanks,
> -Thomas