Is it possible to set a couple of voices (say 3 male and 3 female) that the person interacting with the agent can choose from? Personally, I would like to automate this, maybe have them pick from a dropdown list before they access the url/link to the app/assistant. Or, even tie that dropdown feature to a spreadsheet or something that has specific instructions based on that chosen voice/thing? Is this possible? Or is this another question that's 3 months too early?

@Elastic Absolutely, it's possible to offer a selection of voices for users to choose from before they interact with your voice application. You can implement a dropdown list on your app's interface that allows users to select their preferred voice. This list can be populated with a mix of male and female voices as per your requirement. To automate this, you can tie the dropdown to a backend system or a spreadsheet that maps each selection to specific voice settings. When a user selects a voice, the corresponding settings would be applied to the voice API to ensure that the assistant speaks with the chosen voice. For example, if you're using VAPI, you could have a system where the 'provider' and 'voice' fields are dynamically set based on the user's selection. Since you've mentioned using Play.ht with the voice 'Jennifer', you could add additional voices from Play.ht or other providers to your dropdown list. If you need to implement this feature and require assistance with the technical details or API integration, please let me know, and I can provide more specific guidance or direct you to support for further information.

You can use assitant overrides to let users choose the voice and update the assitant voice right before the call. Check Server event of type request assistant and return the update assistant for the call. https://docs.vapi.ai/server-url/events#retrieving-assistants