We are planning to use sphinx 3 (sphinx3_livedecode) for a simple dialogue system for which we would need to stop the recording automatically at the end of an utterance. Apparently, there is no endpointing implementation in the current sphinx 3 decoder. There is sphinx3_ep which is apparently supposed to return the segmentation in terms of sentences, but we can't get valid result from it so far. We also saw that he Olympus dialogue framework (which uses sphinx 2 and 3) has an audio server which is doing the endpointing, but we did not try yet.
Did anybody face the same issue and found way to get an endpoinying feature?
Cheers!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
We are planning to use sphinx 3 (sphinx3_livedecode) for a simple dialogue system for which we would need to stop the recording automatically at the end of an utterance. Apparently, there is no endpointing implementation in the current sphinx 3 decoder. There is sphinx3_ep which is apparently supposed to return the segmentation in terms of sentences, but we can't get valid result from it so far. We also saw that he Olympus dialogue framework (which uses sphinx 2 and 3) has an audio server which is doing the endpointing, but we did not try yet.
Did anybody face the same issue and found way to get an endpoinying feature?
Cheers!