Voice Assistance - How to integrate voice into your digital transformation roadmap | Publications

Aside from Google, Apple and Amazon we see more and more companies exploring the area of voice. As stated in the previous article “Voice Assistance - Is voice the new wave of digital transformation” the value proposition of voice AI is that it significantly improves customer experiences and in turn allows you to enter in deeper conversations with your customers. In short it is going to be a game-changer. It is a great field to play in, but where do you start? Suppose you discover a great business case for a voice application and you want to develop one yourself. The purpose of this article is to assist you in kick starting your first project.

Build a voice persona

It all starts with a voice persona. In any case, a persona is the face you show to the world, it is the impression that people have of you. It is a topic that has been heavily researched. The book Wired for Speech  shows that we are hard-wired to have biases because of the voice that we hear. Basically within a second of hearing any voice you assign a personality to it, with all kinds of dimensions: age, gender, a local accent that gives you a sense from where they are from, the relationship (my superior, my peer, …). All of this comes through in one second. From that, you make assumptions such as intelligence, ability, and even trustworthiness. All of these factors are milling around in our brains when we are hearing someone's voice. Therefore it becomes absolutely essential that brands design these voices and these personas. Otherwise, listeners are going to design it themselves, and that can be problematic.

To build a voice persona as a brand there are few steps to take:

  1. Make sure to have a deep understanding of the brand and the brand dimensions
  2. Understand the task that the voice needs to do
  3. Who is the user group, what are they like and what are their expectations

This needs to be taken down to some principles of what the persona should be like, in the form of a biography. It will first assist the actors, responsible for recording the speeches, to get into the persona. Secondly, for the person writing the actual application to guarantee that the conversation design is consistent with the characteristics of the persona. The persona creation can be built by any creative expert. Once it is robust you can go ahead designing a conversation. A conversation designer can curate the conversation, defining the flow and its underlying logic in a detailed design specification that represents the complete user experience. 

Build a conversational UI

A conversational user interface, CUI, has the purpose of creating seamless communication between a consumer and a device. To get a little bit more technical, it is a digital interface that allows customers to connect with your business based on principles of real-life human interaction. The core technology used by CUI is Natural Language Processing (NLP), which is a field of artificial intelligence that gives machines the ability to derive meaning from human languages. It is a discipline that focuses on understanding the real intent of a user’s command.

A key question you might ask is whether the ecosystem is already here to link up your users with voice? The answer is yes. To bring your voice experience to life there are end-to-end development tools available in the market that can be called upon by Google, Facebook, Amazon, etc.

Build a voice-interface with Dialogflow

DialogFlow, is a Google-owned platform for natural language processing that allows developers to build conversational UIs. It is designed to model voice interfaces. The tool offers voice recognition and processing as a service. The steps to create a working voice-interface are simple and straightforward:

  1. Training phase: Input a large number of question your customers might ask to train the system
  2. Iterative process: Test it with real users to identify potential bottlenecks and adjust the process accordingly
  3. Go live: Integrate the conversational UI to the platform
  4. Maintenance: Use the build-in analytics tool of Dialogflow to see track the performances and guide you to improve your conversational interface

The solution is not limited to Google’s own speech assistant, the conversational UI can be integrated with a variety of platforms: Facebook Messenger, Skype, Slack, Hangouts Chat, Twitter, Amazon Alexa, etc. Because DialogFlow is developed by Google, the platform is well integrated within Google’s ecosystem. For example giving you access to any Cloud Function, which allow you not only to boost your conversational UI with AI technologies but to scale it on the backend side of things.


publication author dhan claes
AUTHOR
Dhan Claes

| LinkedinThis email address is being protected from spambots. You need JavaScript enabled to view it.

Tags:

Get in touch

Semetis | 44 Rue des Palais, 1030 Brussels - Belgium

+32 (0)2 211 34 50

welcome@semetis.com

Connect with us