Topic: Reduce silence period before processing speech : Politepix

This topic has 4 voices, contains 4 replies, and was last updated by aerialcombat 162 days ago.

Viewing 5 posts - 1 through 5 (of 5 total)

Author	Posts
Author	Posts
June 9, 2011 at 5:44 pm #4114
dave	Hello, I have an app where I want to issue one word commands. However having a 1 second delay before the speech is processed is unacceptable in terms of feedback to the user. Is it possible to adjust this 1000 microsecond delay anywhere in the code? Thanks
June 9, 2011 at 9:36 pm #4115
Halle	Hi Dave, The delay is required in order to prevent cutting off the user mid-statement (for continuous recognition, a period of silence is the only available signal that a statement is complete and ready to be recognized), but you can shorten it for single-word vocabularies in OpenEarsConfig.h under the constant kSecondsOfSilenceToDetect. It’s covered in the docs here: http://www.politepix.com/openears/yourapp/#configoptions
June 13, 2011 at 11:50 pm #4116
alannet	I would like to be able to change the kSecondsOfSilenceToDetect dynamically based on a setup parameter such as a slider in a configuration view. Is that possible? Thanks
June 14, 2011 at 6:43 am #4117
Halle	Since it’s a constant, you’d have to change the way it is set and accessed in ContinuousModel yourself in order to do that.
December 8, 2011 at 6:17 pm #8233
aerialcombat	I had the same question of wanting to change kSecondsOfSilenceToDetect dynamically, but didn’t know how to go about this. If anyone has an answer it would be greatly appreciated.