Ep 14, S3 - Live Accessibility in Action: Behind the Workflow at St. Thomas University - Casey Eakins

This podcast is for city communications teams and video professionals in government. We talk about expanding service delivery with video and streaming, media accessibility, gear, broadcast and streaming workflows, and more. It's all right here on the government video podcast. The government video podcast is brought to you by MediaScribe from Tightrope Media Systems. Tightrope believes accessibility should never be an afterthought. It should be built in from the start. MediaScribe's award winning captioning and audio descriptions is your end to end solution for accessible government video. Visit mediascribe dot ai to start your free trial today. Welcome back to the Government Video Podcast. I'm Michelle Ali Maradi, and I'm your host this week. And I am here with Casey Akins, who is the senior media productions technician at the University of Saint Thomas. So we've been talking a lot about accessibility policy and putting together work plans, but today we're gonna deviate a little bit from that. We're gonna get deep into the workflow aspect of accessibility, especially at a live event in person. So how do you make those events more accessible? What does it look like to implement something like that? And we're gonna really get into the details and find out what it's like a day in the life of implementing some live captioning at your events. So but first, Casey, we wanna get to know you a little bit So tell us about your role. How long have you been at St. Thomas? How did you get into it? Give us Yeah. So I was originally a student at St. Thomas in twenty twenty. I started my freshman year in COVID, which is a whole thing. Yeah. Was really enjoying video production and all that kind of stuff. Joined the student news organization, Tommy Media at the time, now called The Crest, moved past that and then found a love for live sports and production. If you don't know, St. Thomas was the first school to ever go from Division III to Division I in college athletics. It's a very big story. Nice. And this past year, we just finished our fifth year of transition, so we're now a fully fledged member NCAA participant. Can participate in March Madness tournaments, NCAA tournaments of all the likes, so that's really, really cool. But I graduated in May twenty twenty four and then was hired full time after that, so I've been at the university for just over two years now. And you're making a big splash. And I'm making a big splash. Came from another another big interview last week. Yes. I was out over in Atlanta with the sports video group for the college summit presented on the Lee and Penny Anderson Arena, which is our big, bright, shining star on campus now. Brand new. Yes. Multipurpose hockey basketball arena. Seats about four thousand for hockey, five thousand five hundred for basketball. Hundred and eighty three million dollar project. Fully fridge control rooms, production systems, all the likes. But also doing events. Commencements, graduations, some other secret fun stuff coming up this summer that you'll hear about hopefully eventually. So some community stuff. So and community stuff. Yes. Yeah. Absolutely. So it's been a really, really great experience at Saint Thomas. I've loved my time there, and this arena to put on, I'm pretty young still, so to have an arena like that on my resume is unbelievable at my age. Big deal. Yeah. So there were a lot of decisions That you guys had to make when you were putting together, you know, what the production workflow and setup was gonna be like in the stadium. How did accessibility come into the conversation? Yeah. So our our old basketball arena sat about two thousand people. So it's and I wasn't in a lot of these conversations, but it's small enough where we didn't need to meet certain ADA requirements and regulations and all that kind of stuff. Obviously, the new arena meets those regulations, so closed captioning is something we had to add. Never had that before in any of our venues, football, basketball, any of that kind of stuff, as well as accessibility for hearing, so listening packs, all that kind of stuff. So that was a challenging approach to come across, especially something late in the process. Like, were brought up, all these things were made, and, like, we knew we had to have them, but it wasn't until really this past spring that we decided, like, what products we needed, who we were gonna go with. And I also do some Inveni work with Minnesota Twins at Target Field and all that, and they made a big All over the place. All over the place, yeah. But they made a big change this year too for their closed captioning, so having a sonographer on-site, they went with a online server that I think, like, twenty one other MLB stadiums use. So it was unique seeing what they're doing and then how we can take that in. So then that's when we found MediaScribe local company, and that's been a phenomenal workflow for us. So did you have an ideal workflow in mind going into it? Did you have any idea what you would have to do in terms of implementation? So our integrators was Alpha Video Group. They had all the installs of the project and stuff. They'd originally spec some sort of closed captioning card, but it was more of how do we get closed captioning embedded on the video, not how do we actually create closed captions. So that was really on the university side to find, and last April, I went down to NAB, and we had some connections with Tightro Media Systems through Z Play and other products we've had before and saw MediaScribe, at NAB on display and got to get the whole rundown of that, we're like, that's what we want at the arena. One button to hit go, and it just works. It's great. Yeah. And so there's not much else to talk about. No. Am right? You just turn on a button. So, I mean, you guys have a pretty big crew, which a lot of our listeners don't have the luxury of. But talking about just pushing a button, I mean, one person on your crew is operating how many different things? Yeah. So this so, like, a great example is we just finished up eleven high school graduations at the arena, and we were running with a crew of about seven people, one full time, and six students. We put a student in charge of you are starting and stopping the closed captioning in the arena, enabling when it starts and stops, as well as displaying those closed captions, whether it be on our Dectronic video boards, on the phone interface, or we burn those captions into IPTV displays around the concourse and stuff. So if you're walking the concourse, since audio is behind the video a little bit, you can certainly see captions and understand what's going on. We did one student kinda cycle through all those. We did a checklist for them to go through, make sure you're getting the right things in right moments. So it doesn't have to be somebody that's highly trained. Correct. Yes. You can have somebody come in, students and intern. They could be somebody that changes out every single day. Every single day, yeah. I think, yeah, we did eleven graduations. I bet seven different students did that job of push this button, push this thing, push this thing at these specific time intervals, but across, yeah, multiple different events. And I think a lot of our listeners can relate to that part where, like, the person that has to operate this isn't necessarily like, they're not necessarily a video specialist Correct. In any way, so they need it to be very simple. They need it to just like Just work. Yep. Yeah. Yep. So let's talk about what the output is like. You mentioned that you're you're burning captions in in other areas, so like open captions in some areas, and your closed captions are displayed where? So for these commencements, we've got a, like, u shaped ribbon board across our arena with three zones we call our scoreboard zones for hockey and basketball. Obviously, with the commencement, we don't need the scoreboard, so you replace those zones with closed captioning. Okay. It works out great because it's just long enough to get about sixty, seventy characters total in there, which is maybe a little bit long for closed captioning, but for our use case and the fonts we use, trying to keep in the brand of the university. I'm a big brand guy, so I wanna keep everything looking clean and sharp. I'm in marketing, so Yeah. You get it. You get appreciate that. Yeah. Zones looked great. Played a little bit of our own with tweaking with Daktronics as the display provider, so it's a TCP data stream from our Mediascribe server to Daktronics, which I found out is actually faster than the Mediascribe output itself. So whatever back end system they're using to get that data feed out, Daktronics isn't receiving for anything else, which is kinda funny. But tweaking that a little bit on the DAC side, making sure that the lines look good, everything lines up, and it displays properly, worked out, and then, like you said, open captions on our IPTV displays. And then some of the high schools asked for closed captioning on their streams as well. Something like YouTube could just take the audio and auto caption those streams when we're streaming to, I think it's called EDU Vision, some other third party sites that don't have that capability. So us embedding our captions onto our video encoder to send to them really helps them if they don't have to hire out someone else if we already have the captions built on-site. That's that's actually a really good plus for them. Hey, GVP listeners. Before we get back into today's episode, Tightrope Media Systems, the sponsor of Government Video Podcast, has a special offer to help you in your video accessibility journey. Tightrope is the maker of Mediascribe, a platform that makes your government video accessible with automated captioning and audio descriptions. And today, they are offering a thirty day extended evaluation period of Mediascribe Narrate, their solution for automated audio descriptions just for our listeners. So by going to app dot mediascribe dot a I, you can sign up for this offer with code GVP thirty. Now this offer includes training and support and is designed to help you start planning your compliance workflows today. There's no obligation to buy, and this is TypeRoap's commitment to helping teams start figuring out their accessible media workflows right now instead of waiting until the last minute due to budget or procurement limitations or whatever else might be holding you back from getting prepared. MediaScribe Narrate generates audio descriptions for your prerecorded on demand public meeting content. And it's built around how councils and boards actually run. So it prioritizes the most important context like agenda slides, name graphics, and speaker transitions. And that grounding is what helps keep your review steps short. So the accessibility work fits inside a small team's existing schedule rather than adding to it. And solidifying that workflow now instead of waiting until right before the ADA title two deadlines means less to manage later. Tightrope also runs MediaScribe Academy, which is a free resource to anyone working on their content remediation at no cost and regardless of which tools you use to get there. And Tightrope also offers regular educational sessions for MediaScribe users at no added cost. So to redeem this offer, sign up again at app dot mediascribe dot a I with code GVP, like government video podcast thirty, and start building your workflow today. Now back to the episode. You were talking a little bit about speed, so let's talk about latency. Yeah. How how is the latency working out in the space in that, you know, you're doing a live event like graduation. It's kind of important that the person's name is showing up when the person is still there. Yeah. Right? So how is that going? So name's an interesting one. We found that the pace of a lot of name readers Was too slow, so between first name and a last name, there was always a period. So every name was broken up, so we, after our first graduation, we decided to stop captioning names. We don't put graphics on screen, which is something I hope to change next year, but stop capturing names because then it's also, like, for more complicated names, it just doesn't get it quite right. And so we'd rather show no names than show very broken up pieces of names or a misspelled name. Right. Because that's something where you can't necessarily, like, plan ahead Plan ahead anything like that. Yeah. Especially if you've got, yeah, eleven graduations. Each graduation is four hundred to seven hundred students, so trying to feed all of that vocabulary in is just really tough for the system and tough for us in the workflow aspect. But as for latency, traditional one to two seconds I would find on, like, a TV broadcast. I think it's faster than a live photographer. I'll just say that from what I've experienced at least. As long as people can follow along Yep. And it's not, like, noticeable. Right? Yeah. So that's that's good that you're finding that it kinda keeps up with the event. Are you delivering the captions in any other ways? Are you doing, like, mobile device delivery or anything? Yeah. So before all these commencements on our center hung video board, we cycled through the QR code of people at in venue can scan. We also did that for some of the streams so people at home could just pull up on their phone rather than having to pull it up on their TV or the browser. Can be complicated for parents, especially older folks, so it's easier if they just give them the QR code and, like, watch it on this screen, your video's on this screen. Two separate devices, but So in venue cycling the video board, we also have got some signage posted around the arena, so if you're walking around the concourse, you can just quick scan that QR code and have it pulled up. We also do then translations, so we translated for the UST commencements, Spanish, Hmong, Somali, and Chinese, I believe, which is super, super helpful because we have a very diverse student body and just class in general, so that was super helpful from a university side. If that's something they wanted, we're like, yes, easy to implement. And those translations were delivered via the mobile? Root captions. Yeah, via the mobile captions. Yep. So the experience is someone is either in the venue or or at home, and there's a QR code somewhere On the screen, or I guess it's on the screen in your Yeah. On the screen everywhere. Yep. They pick a screen And they scan that, and now they're gonna see either, you know, whatever the original audio is, English probably in this case, and then also, like, the option to view translated captions in as many languages as you As we select. Yeah. Yep. So that's it's great that you're able to, you know, serve the multilingual audiences even better. So it sounds like you don't really have to spend any capacity monitoring this during the event. Not really. No. I think for the commencement stuff, it's really easy to understand. Speakers on stage, if I see words popping up, it's probably working. Yes. I'll I'll follow along the words almost like a script of what I'm hearing and what I'm seeing on our in monitor display or in room display of just the whole transcript. As long as it's lining up, I'm happy. For the sporting events, it's unique in the sense that a traditional, hockey pian outro will draw out names, draw out gold, penalty, all that kind of stuff. A lot of color. A lot of color in there. Yes. A lot of fun with it. So it's unique seeing how MediaScribe takes that in. Sometimes it makes some adjustments based off of like goal. Yeah. It'll just show goal. There's no extra characters or anything like that, which I get accessibility reasons. Yeah. And, like, the extra pizzazz he puts in the voice, and, like, we've got a very Minnesota, PA announcer right over here, which is great, especially for the hockey environment. I can hear that now. Yes. Yeah. But some names and stuff get a little funky in there, and so the arena finished in October twenty fourth twenty twenty five. It was our first hockey game. We weren't allowed in the building until late August, early September to even start playing around with gear, but we had to set up switchers, cameras, replay machines, graphics, IPTV, AV systems. So we kinda just knew closed captioning is if we give it an audio signal, it's just gonna work. So, like, when we started the arena, we were feeding everything from our PA system being music, anything on the Daktronics video boards, anything from our PA, ref mics, all that. We quickly learned music's horrible to feed into there. Even if we give it clean music, MediaScribe will find the profanity words and sometimes display those, so we quickly turn the music off, turn on the profanity filter, solve a lot of problems there. But something I really wanna do next year is implement the vocabulary system and get, like, our at least our home player rosters in there for it to try to get some of those names correct. So take advantage of the ability to Yeah. Tweak it on your end To not just get, like, whatever it's gonna give you Correct. Yeah. no training whatsoever. Yeah. We've we've got a lot of international students, especially in hockey. I think there's, like, four players on our women's teams that are from, like, Switzerland or over there and all that. So their names get a little bit more unique. So if I can at least try on the front end to help the system understand those names, so if a parent parent in their venue sees their kid's name up on screen and sees their caption, thing that the two match, then I'm happy. So it sounds like for before the next graduation, you'll Yes. You'll get a roster Hopefully get a roster. In there Ahead of time and and probably get a better result. Is there anything any advice that you would give to anybody else trying to start up a workflow like this, whether it's the same, you know, level or or something much smaller. I think the biggest thing I learned is being very intentional of what you're feeding your caption server. Like, how we started with feeding everything from the PA system, music, video, all that kind of stuff. We quickly learned we didn't need music. I've been to a lot of I'm a big sports fanatic, so I've been to a lot of sports games in Venue, and I never see closed captioning for any sort of music. But then thinking, like, well, yeah, so there's gonna be a PA announcer, but there's also gonna be an in arena host. There's also gonna be ref mics. Well, you might play a video on the center on video that might have captions embedded in the video already for it, or it might not have captions, so you might as well add that in there. So try to be intentional of what you're feeding the system and understanding viewers either in arena or at home who will be watching that content and what they're gonna wanna read instead of hear. Yeah. And I think it's you mentioned, like, all the different people that would have to be mic'd, and I think that's a conversation that comes up for us a lot. It's, like, trying to understand accuracy, and people don't understand all the things that go into that, right? Correct, Sometimes you can train it if you have the time, but then there's also things like making sure everyone's individually mic'd, and making sure that that sound is actually isolated. Or they're talking to the mic the right way too. Right. Like we had a student last night at our graduation who was standing six inches away from the mic. Well, I'm like, I can't run down there on stage and put the mic in front of her face. Like she's gotta figure that on her own. That's a I think that's another problem that you see a lot in, like, city council meetings because you'll have folks that aren't super mics happy coming up and trying to use them. So I think that's an important thing to try to get ahead of Yes. When you're when you're implementing a solution like that. Or for us, hockey refs. They've got a little mute switch on their pack here and Sometimes they don't Sometimes they don't they sometimes don't even hit the switch. I'm like, you you can tell you're not getting in the arena. You can hear yourself not anywhere else. So it's like, just take that extra step. Some of our facilities folks have talked to the refs after a few times being like, hey, guys. It's just this one button. It's on or off. Try at least talking a little bit before to make sure that you can hear yourself and Yeah. I think that that audio component in addition to, like going a little bit further down, right? To make sure that you're you're not feeding garbage in Yes. So you get the best result on the other end. So I think this is all really great advice. I think it's so impressive what you guys have put together in such a short amount of time, and I'm excited to see, like, how it keeps getting better in season two. Yeah. So I know one of the big things is so we actually have two servers on campus, and they both live in the same rack, both in the arena, but a big thing we advocated for for getting the second server is that if we do, like, board of trustees meetings, we can take that video on our north campus, fiber it all the way down to south campus, and feed it into that closed caption server. Then take the output of that, fiber it back to North Campus, we've got captions for our board of trustees meetings, or if something in one of our other event spaces on North Campus that needs captions. So we were in for sure investing in our future for that kind of workflow as well of, hey. We've got these servers that are gonna sit idle between the months of May to September. We might as well use them for other use cases or weekdays, like weekday weekdays or weekday workdays in the mornings and stuff. Like, we're not playing events at nine am, so board trustees or another event can use that kind of gear and support. It's really, really smart, right? Because you don't have infinite budgets, And so, like, always having that moment where you think about versatility before you implement a tool is also a really important thing to Before we end today, I also was curious, you're talking about, like, advocating for these things. Like, who are the stakeholders? Like, what was that process like for just selecting a product and then getting that proved, like, for folks that are trying to navigate that themselves? Yeah. So I technically work in the IT department at University of St. Thomas, but my biggest partner, I'd say, is athletics. I work a ton of games and all that, but like on my team, one of my colleagues is either the lead or co lead of like the accessibility team in our IT department, So he was a really big advocate for getting something that's really flexible for those external events that aren't athletics related, or which I just found out, like in the last week, I can just drop a video into the MediaScribe server and I'll caption that video. Doesn't have to be a live translation or live captioning. It'll just do it for me, and that's something what we're looking for in other use cases of. If we record something in a studio that can't do that live captioning, dropping it in there, least sending the transcript and putting that into something is really, really helpful for a lot of our other workflows of academic media, for professor, like, classroom content videos, marketing videos, or anything else like that. Yeah. That's a whole other aspect that we haven't even talked about is the ways that you can use transcripts. So do you have people doing that now, or is it still, like, the point of discovery? Like, people don't necessarily know that It's not resource is there? It's not that that resource isn't there, it's just that everyone's kinda doing their own thing right now. We have this other service called Panopto, which is just a video hosting platform that can do captions. It's like my colleague uploads all of his videos to there right now to get captioned and all that kinda stuff, where I'm like, no, we've got something on-site, on prem that can do that. Let's try using this, and there's a different workflow and style for that, so something we're, yeah, slowly figuring out is kind of more designated workflows instead of letting everyone run wild and kind of choose their own Yeah. Yeah. Yeah. That's great. Casey, thank you so much for coming on. Yes. Thanks for having me. And giving us such a, like, detailed view into your world. I really love to hear it. We'll we'll probably have to have you back to get to get a check-in, like, for next season and see how things are see how things are going? Yeah. Absolutely. See if maybe you've gotten some some more buy in for your on prem solution. Thanks so much. Yeah. We'll see you again on the next episode of Government Video Conference.