RTÉ Radio: The Irish public broadcaster wants to lead the AoIP and visual radio conversion

Microphones and speakers sports at RTE 1 Radio Studio

Interview with Michael O’Rourke, former Head of Broadcasting at RTÉ

RTÉ Radio is one of the oldest public radio networks in Europe: born in 1926, RTÉ Radio has undergone every major technological change in broadcasting. Being aware of that, the broadcaster wants to keep being the technological reference of Ireland and that is why is planning a complete renovation towards an AoIP infrastructure and is developing what they define as the “next generation visual radio studio”.

Michael O’ Rourke, Head of Broadcasting at RTÉ, is retiring a few days after our meeting. He tells us the evolution of the station and defines its future in this interview that begins with a chat in which he shares his concern for the next generation of radio listeners…

That’s one of the things that the most part of the people we interview tell us: they don’t know how to connect with the new audiences…

Yes. The challenge is diverse listening habits, right? It’s the same in television. I’m head of broadcasting and technical support services onto Friday, when I’m retiring (laughs). We’re seeing a huge drop-off in live viewing on TV, particularly in that young audience; is kind of scary. I think that radio is in a healthier place than television because you only need one sense and you can listen it on so many devices across so many places. The important thing for us is that we can be discovered on all these new platforms. We are an old radio, we need to be there and we need to be found.

RTE was born in 1926. What have been the main technological changes that have define the station during the last 15-20 years?

The first big one was going to FM stereo. That brought us an enhanced listening experience. Another big step for us was self-editing: we went away from tapes and sound operators to journalists. The next big step was the playout and automation systems. The automations made a big difference, it allowed our stations to transmit 24 hours, whereas previously they would have finished at midnight. The next big step is the end of mechanical devices. Inside the studios, I don’t have as many devices as before. From a support and maintenance point of view, mechanical devices have been dramatically reduced now that we’ve gone to computers and solid state layered systems. The next big thing was the arrival of multi-platforms: we went from long waves, medium wave and FM to all these new platforms. We still have kept long wave: there’s a station that we broadcast into the UK because there’s a lot of Irish people there and, in fact, we’ve been trying to close that long wave transmittal because it’s so inefficient. But leaving that aside, it’s all about the new platforms. The next big thing was the mobile working: MOJO (Mobile Journalism). It is pretty cheap and can be very productive: it’s a standard now in the business. And the last thing for me is the IP world, the connected systems. Before, you put a system and it was kind of stand-alone: all the systems are interlinked now. It is the next big step forward for us.

On air sign radio

Radio technology is constantly changing. What trends do you identify?

Some stats of Ireland: smart speaker ownership is at 11 percent of adults, 15 + in the country. Spotify is 32 percent of adults and smartphone ownership is 82 percent. The challenge for us is to compete on those new platforms and new distribution platforms. At the end of the day content is king, but you must make it suitable for the right platform in the right formats. I have to be discoverable as well. So, one thing is making content, but sometimes you must tailor it for the different platforms, so that target audience can find us and be looking us first: this is one of the things that lead us to visual radio. Why we went into visual radio is not necessary for streaming live programs, but to get those magic moments in the studio. We’ve been doing podcast for a long time, but they don’t share as video shares. That’s why we went into visual radio, which has been quite successful. The other thing is how we make money on these platforms, because everybody else seems to be making money at our expense. You need good analytics of the platforms you’re working on, so gathering and organizing listener data is a high priority for us. We really have to recruit young audiences, if we don’t recruit them, it’s going to be the end of our business. In our main talk radio station, RTE 1, our listenership is quite old. It’s kind of 45+. Nobody can tell us what’s the exact formula for radio good to go forward, but if you do good content, it always sells. We just have to have our brand presence on all these platforms, so we’re not lost in the digital noise.

Sometimes platforms are conservative: if you make a good content tailored for younger audience, probably people on your team would be afraid of losing that 45+ audience…

Yes. The big thing is that we controlled that distribution before. We’ve lost control of that. Now, FM is still about 95 percent of our business, so everybody still listens on FM, but there are all other platforms. There is a little bit of digital noise and these things are made to be a lot more important than they are but, at the same time, we have to compete there, because that’s where the young audience seems to be going. We’re the 23 percent of the national audience because we do good product and making good quality talk radio and good content is expensive. The real challenge is for music stations, because Spotify and things like that can tailor the music for you. Radio music stations are doing a lot more talk because it’s not just good enough to play good music anymore. So it’s a very competitive market there. Our prime pop station has lost a lot of audience over the last number of years because of the diversity of platforms and the competition in that area as well. It doesn’t take much to set up a radio station by a small automation system and “off you go.”

Some of the shows of RTE Radio One are some of the most listened in Ireland. Do you think that technology innovation and audience ratings go hand in hand in radio or, as you said, it is all about the content?

I think that content is king. People will follow content. You have to be competing on technology, but content is more important and quality content doesn’t come cheap. There’s a lot of people putting poor content – because it’s cheap to make. You will keep your listeners if it is good quality content, so it’s more than the technology. In addition, you have to follow your audience. Wherever your audience is, you have to get your content out there and it has to be discoverable as well.

We’re thinking about technologies innovations such as 5.1. Do you think that such innovations are applied to the interests of large audiences or simply they do not care if they listen to 5.1?

I think they’re just fine with the stereo. There have been so many talks about that and we had binaural sound before. It’s a very niche site. Maybe we’re doing special plays, documentaries or dramas.

RTE, as you said, is a public national radio station. How does this affect technological innovation? Do you think that, as public broadcaster, you need to be always up to date with technology?

In Ireland we are seen as technology leaders. I suppose we would have larger budgets than other people. But it goes in cycles, so when a new television station or a new radio station starts up, they tend to have the latest technology. We’re now starting to go through a major technology refresh in the radio side, we’re going completely IP. We can’t always, but we do innovate. We try to collaborate on technology and content with the commercial broadcasters in Ireland. A good example of that was the Irish radio player, a player that we designed in which all the radio stations in Ireland are available on. I think that the reason we did that was to defend radio and Ireland against competition, so we developed together a single player with all local and national radios. That works well.

radio control room at RTE Radio1

We would like to have a wider shot on RTE Radio. Could you describe how the RTE radio studios network is distributed?

At the moment we are moving towards audio over IP. Basically, we have two networks: we’ve got a media network, that’s for our playout system for file transfer. That’s standalone. You’ve got also the corporate network, and then we’ve got as an audio over IP network, which is our distribution network. There are so many different standards of audio over IP: Livewire, Dante, Ravenna… The challenge is to choose what standard and how do we integrate it… We would like to buy our desk and just plug them into our network, but then you lock yourself in just one supplier. So, we’re just looking that technology and which best work. It’s not fully mature yet, but we are moving over: we’re moving away from copper completely and I think that’s across the organization.

What about the studios?

We have basically one studio type with standard equipment, so if you walk into the studio in Dublin and you walk into our one of our regional studios, they will look the same. So from a support point of view, from a training point of view, it’s all one standard. We’re trying to get a supplier that will meet our audio over IP requirements.

So, you have the exact same equipment in every studio…

Yeah, we tried to do that with everything, is what we call SOP (Standard Operation Procedures). If you want to move from a major studio and you move into a studio next door, it would look the same and it can be configured the same. That means, from both an operation point of view and a support point of view, less equipment and less things to train people on. People are familiar with it.

Regarding the AoIP project, do you have an estimate date when do you want to finish this project?

It’s a work in progress. Following the studio part, we’re doing visual radio. We’ve been doing that for two – two-and-a-half years now and we’re looking at the next generation visual radio studio. We want to bring the radio experience to a new level. In other words, we want to have graphics behind them when they’re talking, things like that. We’re looking at the equipment we’re going to put into that. That will help us: whatever desk we select, that should be the benchmark for the next generation of studios. We’re hoping that at the beginning of next year we will have the next generation visual radio studio up and running.

So you have right now two big projects: IP connection and next-gen visual radio studios. Do you have anything else planned for the nearest future?

That’s all. These are building blocks. We have three types of studios with a small production studio or edit booth. That typically will be a small 12 channel mixer and we use radio MAMs or layered system, so you have a small 12 channel mixer, two guests and 1 presenter: it allows our staff to edit interviews offline or even do a small program. In other words, if it’s only a small show, you don’t need a sound operator. If it’s more than two guests or a very busy news show, he’d have a sound operator outside. And the third model of Studio has come to the production studio, which is kind of music or drama. They would use Pro Tools and a Yamaha desk… We would use kind of the best standards from outside from the production world.

RTE 1 Radio Studio

RTE Radio broadcast via web, app, dab +… What difficulties does this imply regarding the distribution of the content?

What we have to do is monitor all those platforms. We have to supply all those platforms… and then the other thing is monitoring and supporting them. That is really the main thing. We’ve contracts with all these suppliers to take our signals. I suppose were responsible up to a certain endpoint and the support of that. The other thing is how to go in the right format. It’s not just about sending audio, you have to send out the metadata. As we go to extra platforms, they’re more demanding with quality metadata.

We want to ask you more about your visual radio project. Do you think that visual radio implies at 24/7 broadcast of the radio station? Would you choose which programs are video-broadcasted?

I think it should be only certain programs. When we started looking visual radio, we were looking kind of a music television type radio: you put in videos and things like that. But we had a national election in the country and that was our first experiment. We did interviews with the leaders of the parties. We had quality interviews so, again, it’s all about the content. You need visual radio in your studio to catch and record all your programs, but whether you should stream those live I don’t think so, it’s only certain programs. We have a program called Drivetime. To make an interesting program you have to have guests in the studio, right? And during the show, you can’t get all guests to come into the studio, so a lot is telephone. Now what we are experimenting is Skype, so if you can’t get somebody in the studio, you do Skype, so they have a visual presence in the studio. Also, some programs just don’t translate into good television, so there’s no point saying we’ve got 24 hours, because it just looks terrible. The other thing is that we don’t want to start adding extra operational cost to the making of a proper program, so we’re using technology automate as much as possible. The system we’re using at the moment is called Phonebox (Editor’s note: The suite is now called “The Bionic Studio”). It’s an English telephone system. Visual radio is a plug into that and it’s interesting because it allows visual radio clippers and publishes that directly from the studio. It also has a component called visual director and another component called Oasis. That allows you to capture your moments and publish them instantly under various platforms. And also, on one window, you can see all your telephone coming in calls coming, you can see our social media coming in and all that, so we can track what’s happening outside very quickly, highlights messages and then publish them. We have integrated our telephone system and our visual radio system and it works very well. And the beauty about that is it’s not an extra system, it’s just an enhanced system, so from a training point of view it’s a very important thing to us. One of the things were asking ourselves is next generation news visual radio studio. Maybe we will need a vision mixer there. The other thing you need… size is important – right? (laughs). So, you need the person who is sitting back to be at least kind of 800 millimeters from the wall to kind of get a depth to the picture, otherwise it looks like a shoe box. We’re designing out new studios with visual in mind. Visual radios are 360 where’s TV is normally 180, so we will use a triangle shape distribution. It is a different experience; it doesn’t follow the traditional rules of television production.

Audio on demand is a reality. You have podcast and it can be listened afterwards. What is the way in which RTE Radio can connect with younger audiences? Is it podcast or is it all about visual radio?

It’s actually both, I think. We have a product called Clipper and it can clip audio or video and generate their own podcast. You’ve got a program and they use the running order to generate clips of each item: the program staff does it themselves. We used to have our archive people, so the day after the program was broadcasted, the archive people go down and write down the metadata and store that in the system. But that’s actually been done live now at the moment: as soon as a clip is finished, it’s published straightaway and that generates a response. So it is a combination of both: quickly getting things both visual and audio. But ideally a visual thing has a bigger impact than an audio thing. Some people might only want to listen back a show or a particular item. As I told you, the figures shows seven million podcast… so podcast is big business for us. Now we got no commercial revenue on them.

One more thing: is your OTT system (podcast) developed in house or are you deploying an external solution?

No, it’s a development in house. That was developed eight years ago, I think now you can buy off-the-shelf system. In fact, we just launched a new video player just before Christmas and it’s completely outsourced, so I imagine that the next generation will be outsourced: we haven’t got there yet. These products are available off-the-shelf and ten years ago they weren’t available. That’s why we developed them.

Microphone in an studio stock image

We were reading about RTÉ and we found out that made a few years ago transition to a private cloud infrastructure nebula called Nebula. Could you tell us more about this? How does it work? Is it is implemented with RTÉ Radio?

Basically it’s an in house system. We virtualized our server system. It’s kind of like going back to a mainframe again. We decided to start the project when we looked around and saw the numbers of physical servers around the place and the cost of maintaining them and licensing them. Now we used build virtualized systems. The next question is what we can virtualize in the cloud and we’re working just on that: there’s a new product called Access Radio and that lends itself to cloud-based systems as well. What that means for the future is that if you want to set up a temporary radio station, it’s very easy just to connect in. You’re doing an OB or something like that, and you can get all of the services on your desktop by connecting to the cloud.

Are you using IP phones for the real-time covering of events or news? How has been the experience so far?

Yes, we use two apps at the moment. So, we use a Tieline codec from Australia, so they use an app called the Tieline app and it dials up a codec. That’s for audio contribution, so you can do record and FTP back, or you can do a live and it’s very reliable. So the other thing that we use for videos is LiveU, that’s for visual journalists as well. We’ve been trying to do that with smart devices, so our journalists have either an iPhone 8 or an iPad Mini. We give them a standard suite of apps: that should allow them to do more stuff in the field. We will reach a point where most journalists will be supplied with this kit as standard, because it’s quite cheap relatively speaking: for about a thousand euros. By using a couple of SIM cards together and very clever algorithms, even in poor enough reception, you can get your signal through. Next thing is the quality of the metadata that comes back with those clips. When somebody shoots something, they also publish a lot of stuff already. GPS coordinates and things that…

That’s where 5G technology appears and that’s my last question. Have you done any test with this technology so far? Do you think that would change the way of doing live-covering?

We haven’t done any tests yet. But the technology looks very promising. Connectivity is the key to support the new IP based broadcasting workflows. Now the challenge I see is the telecom operators. So 5G has fantastic possibilities, but the telecom operators will only put into place what they’ll make money on, so it’ll be interesting to see what they will actually deliver. I know the EBU are working with them to try and get some of the things enabled. That will be interesting, because we’ve seen before from 4G and things that are some they’re there to make money about and the telephone. As we go to more cloud-based off as well as people to do more things from the field and you will need more bandwidth. The big problem with all this is congestion. If a plane crashes or there’s a something big event in a small location, the infrastructure isn’t there to support connectivity when all these journalists arrive to the area. That is always going to be a problem. Satellite is the only way around out, so you give them satellite phones or satellite broadband.

Well, we have talked about everything related with radio and technology. Do you have something to add, something else that you would like to comment on future strategies?

It is difficult in the commercial world to compete with a lot of these multinational companies like on target our audience. Our radio ratings are a system called JNLR and that’s a survey based system. That happens once every three months and it’s used by the commercial agencies. Whereas, there are other technologies that have much more exact science on who is actually listening so they can target their product much better for the advertiser. That is kind of a real challenge to us, because we’ve seen commercial revenue right across all the local radio station stars its falling. 

Cornerstone technolo
TM Broadcast Interna