I built a recogniser in which the speech is some set of commands with specified structure.
Eg: Switch on light one, Switch on light two etc.,
I am using a grammar file for this task. The recognition accuracy is good but the system is giving output even when i say on light one. But i need the output to be displayed only when i say switch on light one. Is it possible with pocketsphinx deocder?
Thanks in advance...
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi
I built a recogniser in which the speech is some set of commands with specified structure.
Eg: Switch on light one, Switch on light two etc.,
I am using a grammar file for this task. The recognition accuracy is good but the system is giving output even when i say on light one. But i need the output to be displayed only when i say switch on light one. Is it possible with pocketsphinx deocder?
Thanks in advance...
You can use keyword spotting mode to search for the commands in continuous speech. This approach is described in our FAQ on the wiki.