Hume AI mission is to ensure that artificial intelligence is built to serve human goals and emotional well-being, giving app builders the tools they need to craft empathic AI experiences.
From creating virtual assistants to powering real-time customer support, Hume AI opens a vast array of possibilities for voice-driven applications.
This plugin uses an external service to provide streaming capabilities.
Just wanted to give you an update which could be useful for others.
When you stop a call and also use the stop speaking option, you can use toolbox to run some JS at the end of that workflow.
if (window.microphoneStream) {
// Stop all audio tracks to turn off the microphone
window.microphoneStream.getTracks().forEach(track => track.stop());
console.log(‘Microphone has been turned off.’);
} else {
console.log(‘No active microphone stream found.’);
}
This turns off the users microphone and it works on both mac and iphones chrome and safari, plus android chrome.
Hi I’m really interested in this plugin. Does it give you the opportunity to write your own prompts for the AI before the conversation? The Hume playground has this as a feature and I just wanted to check if that came across as part of the plugin?
Hi Redvivi, can you explain where in the workflow on the demo you are sending prompts to the Hume EVI? I can’t see from the guidance, or on the demo. What is the name of the database field that contains the prompt text?
Hi there, I’ve got this working reasonably well now, but I can’t figure out how to save individual messages within the conversation to the database. I can see how to make them appear in the repeating group through AllConversations:filtered’s conversationMessages, but because I can’t access the API directly (I think), I can’t access the JSON except as a whole conversation. Is there any solution to this?
by wise:able
