I am currently working on a project that uses the speech(string) node. It works perfectly fine when i say single trigger words that are embedded in the word list (like “replay”). However, when I say a sentence that contains the trigger word, like “I want to replay the video”, it fails to recognize the trigger word.
As it is mainly a problem with the windows speek recognition, can you think of a possible alternative or maybe a workaround for that situation in vvvv?
i found an older post with a similar problem, but without a solution.
Well, as far as I know you can use azure speech recogniser but it takes quite an amount of time to send to a server and get a replay…
This is not implemented and can be implemented using VL…
Maybe there are more libs you can implement using VL, not sure anything out of the box will suite your needs