Podcast: Play in new window | Download
Margaret Downs:
Hi, this is Margaret Downs, Executive Director with Premier Visual Voice.
Mathew Call:
And this is Mathew Call, Director of Technology with Premier Visual Voice, the creators of Caption Companion, and this is your Assistive Technology Update.
John Anderson:
Hello, and welcome to your Assistive Technology Update, a weekly dose of information that keeps you up to date on the latest developments in the field of technology designed to assist individuals with disabilities and special needs. I’m your host, Josh Anderson with the INDATA Project at Easterseals Crossroads in beautiful Indianapolis, Indiana. Welcome to episode 735 of Assistive Technology Update. It is scheduled to be released on June 27th, 2025. On today’s show, we are super excited to welcome back Margaret Downs, Executive Director of Premier Visual Voice and Mathew Call, the Director of Technology for Premier Visual Voice to tell us all about the Caption Companion and give us some great updates on some new certifications that it’s received.
Don’t forget, listeners, we always love hearing from you. If you’ve got a question, a comment, or someone we should have on the show, drop us a line at tech@eastersealscrossroads.org or call our listener line at (317) 721-7124. Thank you so much for listening and let’s go ahead and get on with the show. Listeners today, we welcome Mathew Call to the show and also welcome back Margaret Downs. Now, she was on the show a while back to tell us all about the Caption Companion, and we’re very excited to welcome back Margaret and to welcome Mathew to the show to tell us some exciting updates and learn even more. Margaret, Mathew, welcome to the show!
Margaret Downs:
Thank you.
Mathew Call:
It’s great to be here!
John Anderson:
Yeah, I am really excited to get into talking about the technology and some of the updates, but before we do that, could you tell our listeners a little bit about yourselves?
Margaret Downs:
Sure. I am Executive Director of Premier Visual Voice. We have been established since 2012 and typically, we’ve offered captioning services, so a wide variety of accessibility, communication access services. So recently, we have developed our brand new product, Caption Companion. It’s been around about a year now, and so we’re just excited to talk about the new updates to it and just different things that have gone on in the past year.
John Anderson:
Awesome. Mathew, what about you? Tell our listeners a little bit about yourself.
Mathew Call:
Yeah, sure. Well, I studied Sociology and I graduated back in 2003, but I’d also studied Spanish and I lived in Central America. So while I was working as a Spanish/English interpreter, I actually went back to school for a degree in American Sign Language Interpreting and later became certified because my youngest sister was born deaf. So our whole family kind of used sign language growing up and it seemed like a natural progression for me. And then in 2010, I had a mentor, a buddy of mine who was a ASL interpreter and a Spanish interpreter and a steno writer-
John Anderson:
Wow.
Mathew Call:
… a captioner, and he introduced me to voice writing and I took it and ran with it. It’s basically where you re-speak everything, including punctuation and everything so that your speech recognition software that’s tailored just to your voice, captions accurately. And I loved it. It reminded me a lot of the Spanish/English simultaneous interpreting I was doing, and I never looked back, got certified.
And then for a decade, I worked as Director of CART and Captioning for a company that my older brother started. It was like a one-stop shop for all deaf/hard of hearing communication services. And that’s where the bulk of my experience came from. We did anything accessible, like offline post-production captions, live captioning, ASL interpreting, remote interpreting, video, everything.
And then it was acquired. We saw it as a good thing at the time, and I went over with the acquisition for a period of time. This kind of gets into the history of how Caption Companion came about. I don’t know if I want to go that deep just yet. Should we save that for later in the podcast?
John Anderson:
Well, you know what? Because my next question, I guess was going to be, just for our listeners who haven’t heard before, what is the Caption Companion? So I think it’s a perfect leeway. So yeah, I guess where did the idea for the Caption Companion come from?
Mathew Call:
Okay, yeah. Well see, Margaret and I have worked together for over a decade. We were both remote captioners at the time, and she had her company, we had our company, but we were actually more of a synergistic — we worked with each other. Sometimes we contracted out to her, she contracted out to us. We’ve always had that type of relationship, and we had an issue trying to get good audio feed from inside classrooms for remote captioning, remote CART. And so we used to just rely on whatever the customer’s device was, and that was a mess. No microphone, just use whatever onboard mic is in the device. The internet is spotty. It was a mess.
So we decided to ship them our own microphone. It helped, but it didn’t solve everything. So sometimes, they would use it, sometimes they wouldn’t. And there was still issues with connectivity. And we finally came to the conclusion that unless we send these customers our own hardware, we package it up, make it dead simple where they just tap to connect with a mic. Everything is already paired, everything connected together. And it worked! It really gave us a foundation of success. We just shipped a bunch of those out. Customers didn’t have to think about it. Then they didn’t have to use their own device. They didn’t have to use their own phone or computer. They could to use it for note-taking, like the other students in the classroom.
So when we got acquired, I thought this company would run with that idea. They had no interest in selling hardware. They just tabled it and went in a different direction. And Margaret said, “Well, if they’re not gonna implement that idea, then I will. And she put together a team, and for a couple years, they worked on the interface, beta tested the functionality. And then when, we’ll talk about this later, but when this thing passed a certification exam that only human captioners had been able to pass, that’s when I said, “That’s it. I’m joining Margaret’s team. That’s where I belong.” And that’s how Caption Companion came about.
John Anderson:
Awesome. And we’ll get into the new certification, but I got to save something fun for the end to keep our listeners on. But I guess in kind of talking about the Caption Companion, when I order one, I guess, can you describe the device and what comes with it?
Margaret Downs:
Caption Companion is an eight-inch tablet — style tablet, so it’s very convenient, very versatile, and that’s how it’s designed. The accessories that it comes with, it comes paired with a SmartMic. It’s an external microphone. You can pin it on your shirt, you can hold it. It comes with the lavalier, I think it’s called lavalier?
John Anderson:
Yes. Lavalier, yes, yes.
Margaret Downs:
Yes. It comes with that. Of course, it has its own on board microphone. And then Mathew can talk about the other accessories that are included with the unit. It’s more for connecting to your computer, connecting to, if you’re in a situation where you’re in a conference and maybe you want to pick up the audio from within the conference room, you can do that. So it’s very, very versatile. Can be used just about anywhere. So yeah, he can probably talk better on that.
John Anderson:
Excellent. And then kind of talking about the Caption Companion, does it require Wi-Fi in order to be able to get the captions?
Margaret Downs:
Oh, my gosh.
Mathew Call:
It does not! Right, Margaret?
Margaret Downs:
No.
Mathew Call:
Yeah, that’s the biggest differentiator, is it’s completely offline. All the speech recognition happens on-device. It’s already preloaded, never leaves the device. So yeah, it can work in an airplane, it can work in the subterranean facility!
John Anderson:
Nice. Well, I know-
Mathew Call:
It doesn’t matter.
John Anderson:
I know that’s always been a huge need because you kind of mentioned students, and it seems like every college campus on the planet has some lecture hall that’s in a old bomb shelter from the ’50s or something because-
Margaret Downs:
Yes.
John Anderson:
Yeah, you’re way down there. There’s no internet connectivity, no cell signal, no kind of anything. And I know that getting CART services or any kind of captions has always been a big challenge for anybody down there. So I love that you don’t have to connect it to anything. I suppose that also helps make it a little bit more secure as well, doesn’t it?
Mathew Call:
Absolutely. So I mean, there have been… You know how many acronyms there are, right? Oh, is it COPPA, FERPA, HIPAA, is it all this? All that has to do with data security. ut if the data never leaves the device, it never has to interface with anything related to that organization, it’s siloed, it’s very secure. It doesn’t make sense for us to pursue those kind of certifications because they’re all about how you handle data. But if we never touch the data, we never see it, it’s all on-device itself, then yeah, it definitely makes it more secure.
John Anderson:
Awesome. Awesome. Mathew, Margaret, I know that you just got a new certification, which is probably going to kind of answer this question. How accurate are the captions?
Margaret Downs:
Well, the NCSP, Speech-to-Text Provider Certification by the Global Alliance, Caption Companion took the test and it passed with flying colors, with a 98.56. There are kind of two different tests that we, as captioners take, human captioners. So it’s a more rigorous test than what we typically take in the fact that it accounts for capitalization and punctuation and content. Caption Companion is the first and only ASR product that has passed this test.
John Anderson:
Wow.
Mathew Call:
The NCSP uses a completely different model. When you look at, oh, how accurate is the ASR, the automatic speech recognition from this product or this company or this, this, this. They’re all using what’s called the Word Error Rate, which does not take into account capitalization, punctuation, or meaning. It only takes into account word-for-word accuracy. So that’s kind of the dirty little secret of big tech! That’s how they can claim that they have high accuracy because oh, it doesn’t matter if it completely… just ignore all that stuff and we’ll get high accuracy!
So there’s multiple other models that are starting to get more popular because people are getting savvy to it, like, “Wait a minute. It’s not even that accurate, you claimed it was.” So the NER model is what the Global Alliance uses, and it’s weighted. If you say one word wrong, but it flips the meaning… So you said “can” instead of “can’t”, that’s considered a very heavy error because you just captioned the exact opposite of what was actually spoken. And so, that’s the test that Caption Companion passed, and so far, no other ASR tool has achieved that. Hopefully in the future, more will, and then that will be better for all consumers. But yeah, for now, we can say we’re the first and only.
John Anderson:
That’s super cool to be that accurate and be contained just on the device. So my information isn’t going, I mean, I think of, and you kind of mentioned all the different letters, the HIPAA, FERPA, I can’t remember all the kind of different ones. But if I’m going to the doctor or if it’s kind of a private meeting of any kind. I don’t always — not that that information is always going somewhere bad, but you don’t know where it’s going. So knowing that it’s staying right there on the device and not having to worry about it, and that it’s accurate. Mathew, kind of like you said, “can” and “can’t” have very big things, especially if I’m talking to my lawyer or my doctor or a very important meeting, it’s really nice to know that I’m going to have the correct information actually coming through to me because that can make a big difference in my decisions, and really how I guess I feel about that conversation.
Margaret Downs:
Right, and the way that Caption Companion has been designed and crafted really adds to that accuracy level. So it’s really specifically, it’s the audio, it’s the dedicated device, it’s the versatility. There’s just a lot went into it, into the design, so that can help too.
Mathew Call:
Yeah. Margaret is right. You might think, “Well, I’ve got my own speech-to-text app on my phone or I’ve got this. Why do I need this separate device?” So there’s a time and place for just using your own solution, but anytime you need to have it caption for more than a few seconds, it really is a drag to have to use your own phone or your own laptop or whatever. The points that Margaret brought up are pretty much the reason why we think it passed because captioning accuracy is all about the audio quality, whether you’re using a human captioner or an ASR product. If you don’t have good audio… you could have the best speech engine in the world and it’s going to put out craptions…
Margaret Downs:
Right.
Mathew Call:
… if it doesn’t get a good audio feed.
Margaret Downs:
Right.
Mathew Call:
So we built that in, this thing, the mic it comes with, how it processes audio, how it’s crafted, out of the box, pristine, as good as it can get. And then if you use it correctly, with the included smart accessories, that’s really what shoots it through the roof.
Margaret Downs:
Sets it apart, yep. Yeah.
John Anderson:
Most definitely. And I can tell you in full disclosure, we do have one here in the library and just actually got a second just because it does go out quite a bit, and we want to make sure we have it for folks whenever they need it. Even with the onboard mic, without even using the kind of external mic, we’ve had pretty good success with folks, if you’re just kind of in conversation or kind of close to people or if we’re out doing a demonstration or something like that. But yeah, then when you connect the other mic for if you’re a little further away in a little different environment or anything like that, it does a very good job of picking up the sound and picking up everything and really putting those captions out there so… Mathew, you kind of brought up some other accessories that are available for the Caption Companion. Well, we’ve got some time left. What are some of those?
Mathew Call:
Yeah, so it obviously comes included, every package comes with that wireless SmartMic and the audio cable to connect to external accessories. There’s a charge cable, all that stuff, lanyard, windscreen. Then there’s some optional accessories you can get as a separate purchase. For example, a custom HDMI cable. It doesn’t work with standard off the shelf HDMI cables, but we do sell one that it works with. So you can plug it into a large screen monitor or a projector or use it as a video output, so for large audiences.
There’s a all-in-one protective case that comes with a built-in kickstand, built-in hand strap, built-in shoulder strap. So you can use it anywhere you don’t… The tablet itself comes with a kickstand pre-installed that allows you to prop it up in any orientation. But this case, you don’t actually have to remove the kickstand if you want because the kickstand is so flat, flush against the device, that you just slap the case right on over that. And also there’s a landline telephone adapter you can get so that you can treat it kind of like… to caption telephone calls.
Margaret Downs:
That’s new.
John Anderson:
That’s pretty cool.
Mathew Call:
Yeah, and then there’s… You can also connect it to wireless mics, like we sell 4-channel wireless mic set in case you want multiple people to be wearing a microphone. But there is one caveat. So when it comes to differentiating between speakers, humans are really good at being able to tell when a different person is speaking. Even if we don’t know the name of the person. Like human captioners, we can kind of tease that out and we can indicate that there’s a different person speaking. Most AI researchers thought that would be pretty easy. Since humans can do it, we can just train AI to do it. Wow, has it been the exact opposite!
So yeah, they’ve been working on it for years and years and years, and it’s still somewhat experimental. Even the apps and products that claim they can do it. It’s called speaker segmentation or automatic speaker diarization. You join a Zoom meeting or a Google Meet or whatever, and it’s like, “Well, shoot, look, these captions are smart. It’s indicating when there’s a speaker change and it says on the screen who says each thing.” Yeah, these meeting platforms don’t have to use AI for that, because guess what? Everybody in the meeting has their own mic, their own audio feed. So it’s very easy!
John Anderson:
And their name right there on the meeting so you know?
Margaret Downs:
Right.
Mathew Call:
And their name and their picture.
John Anderson:
Yep, yep. Yeah.
Mathew Call:
Try getting all those people in the same room speaking into the same mic in a Zoom meeting and see how well it distinguishes between who’s speaking. So that’s what we’re dealing with, is single audio feed, multiple people. And even the apps that claim to do that, we’ve tested a lot of them and the way they work is, they listen for a few seconds, try and establish a model for that speaker, then they go back and retroactively stick in a name or a speaker segmentation and it’s rough. It’s very distracting on the screen and sometimes it doesn’t get it right.
Sometimes it changes its mind a little too often and sometimes two or three different people are talking and they have very distinct voices. And the AI thinks it’s all the same person. It just keeps on going, keeps on going. So it’s still experimental. It’s going to be neat when that type of technology matures. We absolutely plan to put that, incorporate that into Caption Companion. But for now, even if you have multiple mics, it all appears as just single text on the screen. There’s no visual distinction when someone different starts talking. But now you know why, because if you had that, you’d probably turn it off.
John Anderson:
Well, and I feel like if I really had to choose between accurate captions or accurately letting me know who’s talking, I probably would rather know what was actually being said-
Margaret Downs:
Right.
John Anderson:
… as opposed to knowing actually who said it. I can always probably, depending on the situation, maybe look up, see who’s actually speaking, look down and be able to get those caption and information. Yeah, I really feel like that’s something that, that again, it’s amazing some of the things that you think as they build AI and stuff like, well, why isn’t this part easy? If you think of the, what is it, the AI art, how bad it used to draw fingers, but I mean, have you ever tried to draw a hand? It’s really hard!
Margaret Downs:
Mm-hmm.
Mathew Call:
It’s not that easy.
John Anderson:
It’s just amazing the things that is very challenging for the artificial intelligence. But hey, again, if it’s giving the correct captions, at least we’re getting the right information, at least we’re getting [inaudible 00:19:48].
Mathew Call:
Oh, yeah, that-
John Anderson:
So yeah, that’s [inaudible 00:19:49].
Mathew Call:
That’s what amazes me! It doesn’t matter. It’s a huge range of speech. It could be the elderly or a child or accented. It’s incredible how accurate it is. Caption Companion is, it’s so versatile that it could be purchased by individual consumers or clients. You can use it for yourself, but also venues, I mean, whether it’s auditoriums or convention centers or any kind of venue. And in fact, it’s not just anecdotal, we actually have some research to back this up.
The CCAA, which is the, they have a website, CCAA.name, N-A-M-E. And that’s the Committee for Communication Access in America. They actually did a large scale survey about public venues within the United States regarding communication technology that was used. And the results were fascinating. I love to read this verbatim. It’s right there. It’s not behind a paywall. You can read their report. If only one technology was to be available at, remember we’re talking about public venues, the first choice of respondents was CART or captions. It’s right there!
So everybody’s got assistive listening device. Everybody’s got, you see on the wall, they have the little ear icon. It’s like, oh, if you need assistive listening, go here or ask for this, whatever. But that was actually the second… Right? Hearing loops were the preferred assistive listening solution but as far as just a single technology, CART or captions. And I think that makes sense if you think about it, because if you don’t have hearing aids, okay. The assistive listening, you plug it in, get some earbuds. Or if your hearing aid isn’t working or if the cochlear implant, you don’t know what program to use or if you weren’t wearing it that day or whatever. There’s a huge range of, it might work, if, if, if.
With captions, as long as you can see the screen, boom, there’s your access. We know there’s not a one size fits all. Obviously, there’s a time and place for every type of assistive technology and every type of accommodation, but culturally Deaf who use ASL, a lot of that population is fluent in English. And then you have the whole range of hearing loss from mild to moderate, severe to profound. And it’s like, captions are almost like a universal language.
Margaret Downs:
Right.
Mathew Call:
It’s like-
Margaret Downs:
Even people who aren’t part of hearing or deaf, I mean, captions help many different people with different challenges. Yeah.
Mathew Call:
Yeah, and there’s tons of research to back that up too.
Margaret Downs:
Right.
Mathew Call:
That it’s beneficial for a whole range of not just disabilities or-
Margaret Downs:
Right.
Mathew Call:
… scenario challenges, but even just limited English proficient language, English language learners. So yeah, we hope that some of these venues who are savvy about putting in all this audio equipment can start putting in captioning, whether it’s a big screen like open captioning for everyone or just individual devices that you check out the audience members if they request it. But captioning should be a standardized way of accommodating communication in venues.
John Anderson:
Nice. Well, we have just a little bit of time left. Could one of you maybe tell me a story of someone’s experience using Caption Companion that maybe sticks with you or sticks out to you?
Margaret Downs:
We get really wonderful feedback from many people. Most recent one, a grandchild contacted us, wondered if Caption Companion would help the grandmother. And we said, “Of course it would!” And they got it and we got this review and the feedback was, “Grandma is just… It’s like a life changer for her. She’s not frustrated any longer when using,” because they had the apps on her phone and just was very frustrating. You get notifications and interruptions. And of course, phone does not have good audio. So that one really sticks out to me, that people can now communicate with the grandma and she’s really pleased with it. So that one was one.
John Anderson:
Love it, love it. Well, if our listeners want to find out more about the Caption Companion or even get one for themselves, what’s a good way for them to do that?
Margaret Downs:
They can go directly to premiervisualvoice.com. There is, at the top of the home page, click Caption Companion. It takes you to another page. And on that page is everything.
John Anderson:
Nice.
Margaret Downs:
All the details about Caption Companion. There are help/demo videos toward the bottom of the page. Oh, we now have a secure cart feature on the website, so people can just click add to cart, place their order. They can choose any optional accessories that they need or want. But yeah, the details are all right there. And we are here 24/7. You call us, you text us, you do a virtual chat on the website. You can schedule a virtual meeting with us. We’re here to help always. Technical support, you purchase one and you’re at an event or something and you’re like, “Oh, this is not working right” or “How do I do this?” Or whatever. You contact us. You will always get a human helping you.
John Anderson:
Awesome. We will put all the links down there in the show notes. Margaret, Mathew, thank you so much for coming on the show. Congratulations on the certification as well. So thanks again for coming on the show-
Margaret Downs:
Great.
John Anderson:
… and telling us all about it.
Margaret Downs:
All right. Thank you.
Mathew Call:
Thank you, Josh. It’s been wonderful.
John Anderson:
Do you have a question about assistive technology? Do you have a suggestion for someone we should interview on Assistive Technology Update? If so, call our listener line at (317) 721-7124. Send us an email at tech@eastersealscrossroads.org, or shoot us a note on Twitter @INDATAproject. Our captions and transcripts for the show are sponsored by the Indiana Telephone Relay Access Corporation or InTRAC. You can find out more about InTRAC at relayindiana.com.
A special thanks to Nikol Prieto for scheduling our amazing guests and making a mess of my schedule. Today’s show is produced, edited, hosted, and fraught over by yours truly. The opinions expressed by our guests are their own and may or may not reflect those of the INDATA Project, Easterseals Crossroads or supporting partners or this host. This was your Assistive Technology Update. I’m Josh Anderson with the INDATA Project at Easterseals Crossroads in beautiful Indianapolis, Indiana. We look forward to seeing you next time. Bye-bye.