This is part of a series on conversational intelligence: where the intelligence is today, and how to use it well in business.
In 2012, I was solving a practical problem for a clinical psychologist. She needed conferencing embedded directly into her website. Simple to use. Frictionless. Secure. Her patients could not be asked to download software, create accounts, or find a meeting link buried in an email. The session had to start with a single click, inside an environment they already trusted.
That was the whole brief. Nothing about artificial intelligence. Nothing about language. Nothing about emotion. Just one client with one real problem.
What followed is the reason I can write this series.
One problem became a product
Solving her problem required building something that did not exist yet in a form anyone could use easily. Browser-based video conferencing was just becoming possible. WebRTC was new. Most of what was available still required downloads, plugins, or third-party accounts. None of that worked for a clinician whose clients were in crisis.
So we built it. Through my previous company, Satya Media Group, that work became MySecureMeeting. A plug-and-play application with voice, video, messaging, and screen sharing, embedded directly inside a website. One click away. No downloads. No accounts.
An archive demo from 2014 is still online if you want to hear how I described the work at the time: MySecureMeeting, One-Click Away Secure Video Conferencing. The voice is the same. The underlying belief is the same. Communication should be simple, direct, and human.
The problem kept getting bigger
Embedded conferencing was the first layer. The next question came quickly. What about people who did not speak the same language? Therapy, legal intake, medical consultation, customer service, and education. In each of these contexts, there were moments when two people needed to understand each other but could not.
Satya Media Group closed in April 2016. The work continued through RealComm Global, which expanded the single-click conferencing capability into something much larger. A WebRTC server. Sentiment analysis. Real-time translation.
It was not perfect. Real-time translation in those years was difficult, resource-intensive, and heavily dependent on cloud infrastructure because the datasets and models were too large to run anywhere else. But it worked well enough to prove something important. The problem was not only language. The problem was that translation, by itself, was not enough to carry meaning between two people.
Words are only part of a conversation. Tone, timing, cadence, and emotional register carry the rest. A translated sentence delivered in a flat, neutral voice is a shadow of the thing the speaker actually said. If the goal is understanding, the technology has to carry more than the words.
That is what the work eventually became. A system that could extract emotional context from spoken language, translate the words, and deliver the translated speech with the original emotion preserved. In 2019, I filed a patent on it. In 2023, it was granted. U.S. Patent 11,587,561 B2.
A word about the term
By conversational intelligence, I mean the ability of a system not only to process a conversation, but to perceive and act on signals inside it. Tone. Pacing. Dialect. Emotional register. Cultural context. Intent. The layers that shape meaning beyond the words themselves.
A system can hold a conversation without understanding what is actually being communicated. The words can be right. The timing can be acceptable. The customer can still walk away feeling that something was off because the system was responding to language rather than to the person using it.
What these systems can perceive is often described more narrowly than the underlying technology supports. That gap is part of why this series exists.
The conversation is not the breakthrough. The intelligence inside the conversation is.
Why this series
The technology I worked on an early version of is now central to how voice AI, sentiment analysis, and conversational platforms are being developed across the industry. Tools that analyze emotions, translate in real time, clone voices, and track sentiment are already built into the software most businesses use every day. Customer relationship platforms. Call center systems. Customer support tools. Marketing automation.
Most business owners do not realize how deep their capabilities already run, and what that means for the choices before them.
The work has continued. What I am building now draws on the same architecture but runs in a different environment than what was possible in 2019.
Over the next several posts, I am going to walk through what I learned building early conversational intelligence systems and what that experience tells me about how to use these tools well in a business today.
Some of the posts will be practical. How to add a second language to a primarily English business without starting over. How to think about voice AI that represents your company. How to add real-time translation to customer-facing interactions without losing the warmth that keeps customers coming back.
Some of the posts will go deeper into what the technology actually measures, and what it should never pretend to be.
The thread running through all of it
Communication is not just words. It never has been. Every layer I have worked on, from embedded video to real-time translation to emotion-aware systems, has been an attempt to keep the human in the exchange. To reduce friction without flattening meaning. To carry intent across language, across distance, across the medium itself.
The series starts here, in 2012, with one client who needed something simple. The rest follows from that.
About Mary Lee Weir
Mary Lee Weir has been building websites for 27 years and digital products in 7 countries. She holds U.S. Patent 11,587,561 B2 for a communication system and method of extracting emotion data during translations, and continues to build in the conversational intelligence space. She runs Vero Web Consulting in Vero Beach, Florida, and founded Belize Web and Information Systems at home in Belize to serve Belizean businesses. She writes about AI, search, and the practical realities of building for the web at maryleeweir.com.
If any of this is useful
Book a 60-minute strategy call ($250) to work through how any of this applies to your specific business. Or start with a free 15-minute intro to see whether a longer conversation makes sense.
Conversational Intelligence Series
- Conversational Intelligence: How It Started
- Why Friction Was the Real Problem
- When Words Were Not Enough
- What Sentiment Analysis Became
- What the Patent Actually Covers
- What AI Can Perceive
- Where Emotion-Aware AI Stops
- Cloud Before the Edge
- How to Add a Second Language
- Voice AI for Your Business
- Monitoring Versus Understanding
- What Comes Next
