Hello everyone,
I am implementing a Web voice agent and would like to enable quick actions for the user.
During the call, I want to display options on the screen so the user can select one immediately. This would allow the end user to skip the agent’s full explanation and move directly to the next step.
How could I achieve this? I might have missed it, but I haven’t found any endpoint in the documentation that allows for mixing text/UI elements into a voice flow.
Thank you!
Retell doesn’t natively support mixing text/UI input directly into an active voice call.(1)
However, you can work around this by running a Chat Agent alongside your voice agent. You would convert your voice agent into a Chat Agent in Retell, embed the chat widget using their script, and users can then type or talk using the same conversation flow.(1)
For implementing the web voice interface itself, you can use the JavaScript Web SDK to build your custom UI with quick action buttons.(2) The SDK allows you to implement your own interface elements around the voice call.(2)
While there isn’t a documented endpoint for injecting UI elements directly into the voice stream, you could potentially:
The JavaScript Web SDK is available at: https://github.com/RetellAI/retell-client-js-sdk[(2)](https://discord.com/channels/1186424376300679209/1301491212691312711)
If you need more specific guidance on implementing this pattern, I’d recommend:
Would you like a general answer based on my knowledge instead?