Video: How to Interpret the Transaction in Every Conversation
Learn more about AI and intelligent agent design at the next SpeechTEK conference.
Watch Anthony Scodary's complete keynote from SpeechTEK 2019, How Companies are Partnering with Conversational Machines, in the SpeechTEK Video Portal.
Interested in attending a SpeechTEK event? Visit SpeechTEK.com
to sign up for conference alerts, discounts, and more.
Read the complete transcript of this clip:
Anthony Scodary: All speech is transactional. And I don't just mean necessarily financial transactions but anytime you say anything, there's a purpose behind why you said that, or at least hopefully. And the goal here is to understand every time when we're looking at a conversation whether we're analyzing it in analytics or we're trying to automate it, there is a purpose behind what we are saying and what is being said to us. And I want to look at kind of the fundamental units of these transactions that make up full conversations.
So, a simple transaction: One person has something and another person has something, and they exchange it. In a conversational context, in a call center a caller might have something that they need an agent has it, and then they reach a conclusion. In a more complex scenario, both sides have information that they need from the other and only when they all have exchanged what they needed can the call reach some sort of resolution. I want to look at the taxonomy, the kind of fundamental types of animals that you see in a conversation. These are called dialogue acts that make up this kind of zoo of fundamental dialogue components that form a conversation between people.
Here's the ones that we recognize in a conversation, when we are analyzing it or when we are participating in it in the form of autonomy.
[call center recording starts]
[Male Representative] How can I help you?
[Female Representative] Have a good day. Then it looks like you're all set up. Is there anything else I can help you with today? I'm so sorry Mr. Jones, but we are not the Westin. We are actually Hotels Worldwide. I'm so sorry to hear that, sir. May I have your name?
[Male Representative] One, two, four, eight... D, O, O, K, yep, pull you right up here, and I see that...
[Female Representative] Did you call it in? Or did you you use the the online service? Check-in. Okay, we have a room with two queen beds available. Will that suit you?
[call center recording ends]
Anthony Scodary: So this is a conversation being analyzed live in our analytics product, Sift. When we're trying to figure out what the purpose of each dialogue act is--oftentimes multiple purposes per sentence--you have to look both at a small context and at the context of the conversation as a whole to figure out why someone is saying what they're saying what the purpose of it is, what the subtext is. You also have to look very closely at every word and the way the word is said, both the semantics and the persotics of what they're saying.
If you want to be able to piece together why is this conversation happening? How is it going? Is it reaching a resolution? What are the goals of both speakers? Are promises being made? Are questions being asked? Are answers being given? In addition to the kind of more mechanistic black-and-white aspects of these speech transactions one of the things that conversational speech different than text or chat, is that there is a lot of emotion conveyed that add a lot of color to the conversation. And a large amount of information that we exchange when we talk is in how we say things and not just what we say.
Humans evolved emotion in a social context and a large amount of reasons that we express emotion even though it's an internal state, is that it's a form of social communication that is actually about half of the information in conversational speech. And if you wanna understand what something means and a lot of times when people are calling call centers they have a lot of emotions, you need to understand this just as much as you need to understand the kind of more clinical transactional parts of the conversation.
Allstate Conversational Designer Katie Lower outlines working models for assessing the viability of a conversational interface with multiple teams within an organization in this clip from her presentation at SpeechTEK 2019.
Allstate Conversational Designer Katie Lower defines the customer journey map as a visualization of the customer's process and explains why it's valuable in this clip from her presentation at SpeechTEK 2019.
Grand Studio Lead Designer Diana Deibel discusses the ethical implications of speech UIs and remaining cognizant of the inherent human elements of speech and conversation in this clip from her presentation at SpeechTEK 2019.
Grand Studio Lead Designer Diana Deibel discusses multiple approaches to making VUI design transparent--the Google vs. Alexa, system-initiated vs. user-initiated--in this clip from her presentation at SpeechTEK 2019.
Pindrop Director of Product Marketing Ben Cunningham discusses best practices for voice authentication in IVR design in this clip from his panel at SpeechTEK 2019.
Gridspace Co-Founder and Co-Head of Engineering Anthony Scodary demonstrates Grace, Gridspace's new automonous call center agent, in this clip from his keynote at SpeechTEK 2019.
Orion Labs Head of Product Ellen Juhlin and Voicea CMO Cory Treffiletti discuss persisting challenges in speech-to-text, AI identifying intent, user expectations, and more in enterprise speech tech applications in this clip from their panel at SpeechTEK 2019.
451 Research Senior Analyst Raul Castanon discusses new findings of a recent survey on speech technology adoption in the enterprise and how adoption of devices in the consumer space have impacted enterprise adoption in this clip from his panel at SpeechTEK 2019.
Grand Studio Lead Designer Diana Deibel discusses best practices for culturally inclusive access in voice UI design in this clip from her presentation at SpeechTEK 2019.
Conversational Technologies Principal Deborah Dahl lays out a plan for making more virtual assistants more effective in this clip from her keynote at SpeechTEK 2019.
Conversational Technologies Principal Deborah Dahl discusses the state of the art for the three pillars of conversational systems in this clip from her keynote at SpeechTEK 2019.
Conversational Technologies Principal Deborah Dahl explains how more targeted enterprise knowledge could make VAs more effective in organizations in this clip from her keynote at SpeechTEK 2019.
Nuance Communications' Roanne Levitt delineates the differences between text-dependent and text-independent biometrics and what the advent of text-independent means for IVR applications in this clip from SpeechTEK 2019.