Conversational AI should be in every developer toolbox. Here's why.
Why think about conversational voice technologies in the first place
Where is conversational voice AI today
- Chatbots. Mainly simple, often multiple-choice apps. Great for taking an order or routing a customer to the right support agent. I felt I had to mention, though the focus of this article is on voice.
- Basic voice user interfaces - Alexa Skills, Siri assistant and such. They are great for fulfilling a single function, for example, tell you the weather or play a song. As these are command-response interactions, they do not aim to pass the Turing test and they are limited in scope of application.
- Voice AI business automation. Call center automation from simple voice (not touch tone) menus to more complex outbound applications. Some of these applications (mainly outbound today) do aim to pass the Turing test and oftentimes do pass it.
What will adding conversational AI to your toolbox give you, as a developer?
Voice user interfaces
- If your app is designed to be used on the go (walking, running, driving), for example, it’s a navigation app, a music app, etc.
- If there is a use case in which your app runs in the background, while the user switches to another app.
Automated telephone conversations
- Call a user who abandoned their shopping cart and ask if they need any help completing the purchase or have some questions they need answered.
- Call a user to verify some details or call them up as a part of an onboarding process.
- Set up a fully automated customer service line for your product. You can take calls through telephony or in-app using voice over GRPC.
- Call back a newly generated lead in less than a minute to convert them into a user.
- Build a voice Discord bot that can do pretty much anything you can think of.
- Literally build a replacement for a call center agent and impress your boss at the big bank.