You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Feb 9, 2023. It is now read-only.
In my project, I need to create a conversational bot, something like this:
Bot, put an alarm for me
OK, when will it be?
7.00 AM will be fine
I want to do it with the hotword "bot" (without button trigger), but currently the transcript must include hotwords in order to be recognized. Is there any way to call the recognizer to immediately listens when called while still being able to listen to the hotword for initializing the conversation?
The text was updated successfully, but these errors were encountered:
We don't have a way to do this ("follow-on queries" in the language of the Assistant SDK) with Cloud Speech or with Assistant gRPC at the moment. With the Assistant Library, it will only work for built-in follow-on queries (like "OK Google, hello" ... "Hi, what can I do for you?" ... "What's the weather?").
With Cloud Speech hotwording, you could try enabling and disabling the hotword depending on the previous query. Currently you'd have to access the private member (eg recognizer._hotwords = []) but if it's useful I'd accept a pull request to add an immediate=True parameter to recognizer.recognize().
In my project, I need to create a conversational bot, something like this:
I want to do it with the hotword "bot" (without button trigger), but currently the transcript must include hotwords in order to be recognized. Is there any way to call the recognizer to immediately listens when called while still being able to listen to the hotword for initializing the conversation?
The text was updated successfully, but these errors were encountered: