"My name is... my name is...": A Linguistic Map for Voice Agents — Midam Kim, ServiceNow

Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.

Original language: English

Original video: “My name is… my name is…”: A Linguistic Map for Voice Agents — Midam Kim, ServiceNow · Channel: AI Engineer · 14:44 · Watch on YouTube

Topics: AI agents, Multimodal models

Summary

Midam Kim presents a linguistic framework for understanding why voice agents fail, using a frustrating interaction with a bot as an example. She describes voice communication as a joint activity involving listening, speaking, sounds, words, interaction, and a shared mental model. These components are interdependent and unfold over time, so a successful agent must adapt to the user and maintain context rather than operate as a simple pipeline. Kim recommends using the framework to diagnose systems and highlights the business impact of linguistic quality, including user frustration, task failures, and escalations.

Key points

Human communication is a joint activity, and voice agents should be designed to participate in that back-and-forth.
The example call fails through speech recognition errors, English-centric pronunciation, interruptions, and a lack of interactive clarification.
The framework covers listening and speaking across sounds, words, interaction, and mental models.
The framework’s components are interdependent: improvements to sounds and words must also account for interaction and task completion.
Spoken sounds, words, and interaction disappear over time, while the user’s mental model continues to develop.
Voice agents need suitable ASR and TTS, curated vocabulary, effective turn detection and turn-taking, emotion handling, and context retention.
Agent behavior should adapt dynamically to different users and changing conditions during a call.
Kim points to ServiceNow’s Eva bench as an end-to-end benchmark for diagnosing voice agents.
Linguistic quality has business implications, including user frustration, failed tasks, live-agent escalation, abandoned calls, and silent failures.

Chapters

Start	Chapter	Summary
00:01	A voice agent failure	Kim introduces her background and recounts a call in which a bot mishears her name, interrupts her, and fails to help with an account number.
03:17	Voice communication as a joint activity	She frames conversation as a back-and-forth process and maps the call’s failures to listening, speaking, and shared understanding.
06:30	The linguistic framework	Kim lays out the sounds, words, interaction, and mental-model components across listening and speaking, emphasizing that they depend on one another.
09:11	Designing adaptive voice agents	She explains why user context and mental models matter over time and outlines capabilities agents need to serve users effectively.
12:28	Diagnosis, business impact, and language change	Kim shares the framework’s takeaways, mentions Eva bench, and asks whether agents can adapt as users and language change.

Quotes

human communication is a joint activity — 03:17

these components are interdependent, not separate — 08:10

only the user’s mental model is the thing that’s left, and that matters. — 09:45

voice AI is a joint activity between the bot and the user, not just a pipeline. — 12:28

Speakers are always adapting. — 14:07

Concepts and entities

Midam Kim, ServiceNow, linguistic framework, Voice AI, ASR, STT, TTS, Eva bench, turn-taking, mental model

Full transcript (original language)

00:01 [music] >> Okay, hello everyone. So, my name is Midam Kim. I am an ML engineer from ServiceNow and I’ll be talking about a linguistic framework for voice AI.

00:33 So, quick background of me so you know where I’m coming from. Like I said, I’m an ML engineer at ServiceNow, but I’m also a researcher, lifelong researcher, of speech communication in the wild. So, my motto is doing linguistics and what I’m going to be doing today is to hand you that lens of linguistics. So, have you experienced voice AI failures?

01:05 Yeah, like everyone. >> [laughter] >> So, I’m going to introduce an example that I experienced myself. So, the bot asked me, “Could you please spell your first name?” And then I slowly start to spell my name. Yes, it is m i d a m. And the bot says, “Confirming with you, is it m i d a n?” And then I say, “No, it is m i d a m.”

01:41 Um and the bot says, “Thank you for your correction. Happy to help you today, Madam.” And I then I get slightly annoyed, more annoyed, because my name is Midam, not Madam. And then it asked me about, “Now, what is your account number? And then, I start start getting confused. What is that account number thing? And then, I try to find uh information about that. So, which one? Um it must be and I start

02:16 uh slowly start spelling the account number. So, it is A X 4 5 1. And then, I take time because I’m not used to reading this strange number. And then, the bot cuts me off. And then, it says, I couldn’t find your record. And then, without even trying, it asked me to repeat that again. Can you please repeat that? And then, I get super annoyed and then, I can say, can I talk

02:46 to a person? I just don’t want to deal with you anymore. So, this is a very typical pattern of voice AI, unfortunately, at this point. So, I just want to navigate how we can solve this problem with linguistics. So, voice AI is booming. But users are still often preferring human agents over voice agents. How can we mitigate this issue?

03:17 But in the first place, what are the actual problems? So, I think we can think about a fundamental frame framework to understand this into an architecture of voice AI, which is called linguistics. So, as all of us already know, human communication is a joint activity, like the thing that we’re doing right now. So, I give you my sounds and words. You hear them.

03:49 And then, if it is a conversation, you’re going to give me your sounds and your words. And then this is going back and forth through interaction. And then in this process, we’re continuously processing and updating our mental models. So that’s a joint activity for human communication. And I would like to say in the voice AI human communication,

04:20 it also has to be a joint activity like this. Because that’s the only thing that we know about human communication as a human being. We have been evolving thousands of years as communicators, and this is what we know. So we expect the same thing to bots. So let me go over the failure scene of my call with the voice agent in this framework. So you see there’s listen

04:52 and speak for each party. So I start spelling my first name. And then the bot did not hear that the difference between M and N correctly, so it’s an SCT failure in the listening level. And then the TTS applies only English-centric reading rules to my name, M I D A M, would read it as Midam in the American English version.

05:25 So I’m confused, but at this time I’m kind of generous because that happens a lot even with human beings. So I’m okay. But then when it brought brought up account number thing because I don’t know what that is, I’m confused again. But I’m adaptive, I can find I can look for it. So I found the number, start reading it, but the STT did not recognize the word unit correctly, so it cuts me off, and uh

05:58 uh finally, it’s uh eventually talked over me. So, I get really irritated. And then, when it asked me for the repetition of the same information, and then, it is clear that the spot is not tracking the mental model with me. And then, very rudely, it’s uh does not even try interactive clarification, which is a common strategy by human beings. So, I don’t want to deal with this anymore, so I say, “Can I talk to a

06:30 person?” So, let’s go over the uh the framework again. So, the these are the linguistic components that are expected and well maintained in human-to-human voi- uh uh conversation. So, there are listening channels, a listening channel and speaking channel, and there are different components like sounds, words, interaction, and mental model. So, the first component is, does the bot recognize the user’s speech well? And all of these technical terms

07:04 uh will fall under this. And then, there was there’s going to be this second component, which is words in the listening channel. So, does the bot understand the user’s words? And then, the third one is, does the bot wait until the right timing to for its turn? It’s about It’s going to be about uh listening channel interaction. And then, uh the last part is mental model. So, does the bot understand the user’s intention in the listening part?

07:38 And then, we can also go to the speaking channel, so it’s going to be about pronunciation for the sound. And also there’s about understand the the words users are uh there’s about choose the words the user can understand. And in the interaction part, there’s about speak with the right timing. And lastly, there’s about speak with the information the user actually need. So, there are a lot of engineering or linguistic or cognitive science terms

08:10 that are in here that that are here. Um you can see now see that all of those have their right spots in this linguistic framework. And importantly, these components are interdependent, not separate or uh independent from each other. They’re interdependent and they’re aligned. So, when you want to do good things about sounds, you have to think about words level. And then when you want to do good things

08:41 about these sounds and words, you also have to uh account for interaction, so turn taking or turn detection. And then finally, you want to uh have good uh task completion, which is the goal of these mental model uh layer. Then you have to have all of these. Without all of those, without any of those, any of those components, your voice agent will fail. And then finally,

09:11 uh it has to be well aligned. All of these have to be well aligned. And additionally, you have to keep your mind keep in mind that this is happening on the timeline. What I mean by that is it is silently tracked. Unlike in chat, in chat you see the history of what was said uh as text. But in voice agent experience, uh, you say something, and the bot says something, you go back and forth,

09:45 and then see, all these waveforms, the air via the vibration in the, uh, in the air, they’re all gone. And only the user’s mental model is the thing that’s left, and that matters. So, sounds, words, interactions vanish the moment they’re spoken, but the mental model proceeds and grows over the timeline. So, this is what you have to target for user satisfaction.

10:16 And then, what can we do for the bot to meet the standard of the user? So, what we can do, uh, would include, of course, choosing good ASR models or configurations and do some post-processing, uh, choosing good TTS models, configurations, and pre-processing, and, uh, carefully curate the vocabulary that can be shared between the bot and the user, and do good job of a turn-to-turn

10:48 detection, latency, and turn-taking. Um, and very importantly, we have to, it would be great if we can do good emotion detection and handling, and context retention, and by context, what I mean is context about all of these. And importantly, uh, it has to be dynamic because things are always changing, uh, throughout over the course of the call. So, we would have to do this management dynamically along the timeline

11:21 for different kinds of people. So, kids or different kinds of people like these will have different expectations that we have to satisfy. Uh, not just when they’re happy, but also when they’re not happy. So, only then you can pursue a dynamic and truly scalable orchestration of voice AI. So, it’s a very difficult job to do. We always say that voice is the most natural way of communication, but it is

11:53 not actually not easy. Behind the scene, it is thanks to this linguistic orchestration. When your bot is not good at it, it’s a catastrophic failure. Um, so paying attention to this linguistic framework would have lots of business implications because then you can uh decrease all of these user frustration, task failures, live agent escalation, or abandoned calls, or silent failures.

12:28 So, in ServiceNow, we have made a a good uh benchmark end-to-end benchmark called Eva bench. So, you can try that to diagnose your voice agent’s uh status. Um, key takeaways. So, voice AI is a joint activity between the bot and the user, not just a pipeline. And we must serve users’ needs in multiple layers real time.

12:59 It’s not that I have given you a fix today because there’s nothing like that. It just uh the fix is in you and your system. But, what I have given you is today is the linguistic framework you can try to diagnose your system and to build your system upon. You can try Eva, but also you can learn linguistics and hire linguists. Um, another thing I want to remind you of is that business implications are

13:32 linguistic implications and vice versa in this voice AI scene. Because voice is fundamentally a linguistic and very human and cognitive experience. I would like to ask you a longer term question. Speakers adapt. So, I I’m pretty sure that in this talk in my talk with you guys today, you have learned something about me, about my speaking style, what kind of accents I speak, what kind of words I’m using. So,

14:07 next time I see you guys in person, you would find it more comfortable to talk to me because you have paid attention to me. Right? So, speakers are always adapting. So, the user will be adapting to your voice agent throughout the call. So, is your system ready for them to use you better, use it your voice agent better the next time? And language is always change. So, is your voice agent ready for language change in 1 year or 6 months even?

14:44 So, thank you. >> [applause]