Services Plugins FAQs

Help with Speech Recognition Plugin

Plugins
#1

Hi there,

I am working on an app in bubble.io using a primary input method of the speech recognition plugin, although, for the life of me and my partners, we cannot figure out how to get it to function even just printing basic speech, any help would be greatly appreciated! Even to just get what is being said printed as a text element would be amazing!

Thanks for the Help
Rocco

#2

Hi @turanor,
Thanks for reaching out and purchase of the “Voice Recognition with Action” plugin.

We tested the plugin functionality on our side, and no error was found.

We kindly recommend doublechecking all settings that you have set within your app, especially paying attention to the following moments:

  • check the microphone input settings on your device;
  • check if at “Live Demo” / your app page loading, you get a pop-up notification that is warning about granting access to the microphone? It might be possible that the browser itself has restricted the access of the application to the microphone.
  • check and compare the workflows and settings from the “Live Demo” page with the same on your app (timing within the workflows, etc)

Hope it helps. In case the same issue persists or any other help is required, please let me know and provide the information on your plugin setup.

Regards,

#3

Looking through the example, there’s a lot of unecceary workflows built around specific actions in the demo. Is there a way to simply start the plugin on page load, and have it print what is being said in a text box? Rather than evaluate for all the actions necessary in the demo.

Thank you!
Best, Stay Safe
Rocco

#4

Hello @turanor,

Thank you for your reply and provided feedback.
In order to increase the usability of the plugin, we will add more information to the plugin description in nearest future

Most of the workflows presented on the plugin demo page are intended to show the plugin’s multiple functionalities. Considering that they are only for demonstration purposes, it means that it is not required to recreate all of them on your app.

Coming back to your question, you may build your workflow, basing on the custom states & actions, described on the plugin page: https://docs.zeroqode.com/plugins/voice-recognition-with-actions-for-bubble-apps-plugin

For example, a new workflow may start with the following: https://prnt.sc/xyuc7w

You also may copy a certain workflow, and paste it on your app, to reproduce/modify the action it may do, for example:

  1. https://prnt.sc/xytuvp

  2. https://prnt.sc/xytvzw

Hope it helps.

Regards,

#5

Hey guys, with this plugin - is there any privacy implications I need to make clear to users if I have the plugin listening from page load for trigger words?

Does the plugin store these recordings anywhere or is it just listening for the trigger wordds

Thanks!
Paul

#6

Hi @paudavis363, thanks for reaching out.

For the voice recognition functionality, the plugin is using a lightweight js library for voice recognition named annyang, so, it is not storing any record in the app database or elsewhere.

Following this, there is no need to inform your app users regarding any privacy issues.

Please find more info about annyang here: https://www.talater.com/annyang/ , and the FAQ available here: https://github.com/TalAter/annyang/blob/master/docs/FAQ.md#can-annyang-be-used-to-capture-the-full-text-spoken-by-the-user

Hope it helps.

Regards,