Continuous Speech Recognition

EddyVerbruggen / nativescript-speech-recognition

:speech_balloon: Speech to text, using the awesome engines readily available on the device.

Other

90 stars 23 forks source link

Continuous Speech Recognition #22

Open tylerablake opened 6 years ago

tylerablake commented 6 years ago

More of a question than a bug I think but hopefully you can help.

Is there any way to get the plugin to do continuous speech recognition listening for a keyword or only able to operate by push to talk for now? How should I handle enable/disable speech recognition while device is trying to play audio then re-enable after device response is done?

I am hoping to use this plugin in an app but having trouble getting it to run on a device while using NativeScript-texttospeech plugin to respond after keywords are recognized but after the texttospeech plugin starts responding, the app crashes. I believe due to input/output busses fighting over each other..

Thanks for your help!

Clifte commented 6 years ago

I have the very same problem. I worder if is it possible to configure the silent period for both platfforms. About the nativescript-texttospeech I am not using.

nicolasfguillaume commented 4 years ago

Same problem here, I would like my app to react to a specific keyword. Thanks a lot for your great work!

EDIT: after doing more research, continuous speech recognition does not seem to be allowed natively neither on apps nor on web browser apps. I see two more options:

streaming continuously to a third-party online speech recognition service (cons: may drain the battery)
train a machine learning model and run it on the device, for example, see: https://www.npmjs.com/package/@tensorflow-models/speech-commands (cons: require knowledge in data science)

liamcharmer commented 3 years ago

Any further work with this?? At the moment I have the continuous voice and then based on the string then check if a certain word is matched and if it is the case then run commands. However this is very buggy...