There is just a couple of weeks to go before submissions close on the government’s proposals paper about Radio Prominence on Smart Speakers.
Last year the government examined prominence for free to air Australian TV Networks, resulting in a range of requirements for smart tvs and other devices to ensure access to broadcast tv is at the front of the tv menu, not buried behind other apps and options on the home screen. The new requirements are now enshrined in the Prominence and Anti-siphoning Act 2024.
Radio was not included in that legislation, but is now likely to get its own prominence legislation to guarantee Australian free to air broadcast radio remains easy to find and access on smart speakers.
A strong outcome is needed to ensure that Australian radio has the ability to adapt and grow in the new digital audio environment in the face of international competition and multinational control of the pipes and the commands that deliver audio to your phone and smart speaker.
The path between a radio transmitter and the listening audience is simple: studio > to transmitter > to receiver. The broadcast company sometimes owns its own transmitters or contracts transmission out to a known transmission supply company, such as BAI or TxA in Australia. The contractual and quality assurance relationship between the broadcaster and transmission supply company is a direct one. Whether one or one million people listen, the quality of the signal remains the same.
Contrast that with the path to a smart speaker, which is roughly:
> studio to media encoder
> encoder (or media hub company) to internet
> internet to multiple audio delivery companies (ABC Listen, CRA RadioApp, iHeart, LiSTNR, TuneIn Radio, Apple Music, Amazon Music, Apple, Audacy and others)
> delivery company to receiver device, through internet or cellphone (variable bandwidth)
> receiver app activation via voice assistant on smart speaker, phone, computer or smart tv
There are some major considerations as your radio stream passes through the various steps above:
- Will the encoding and delivery companies maintain the integrity of the signal at the front end, and if they don’t can you do anything about it?
- How will the companies in the various steps adapt to changing bandwidth and buffering?
- How will your receiving app and your phone/computer react to changing bandwidth, will they downgrade the signal, buffer it, decrease the EQ bandwidth or delay the signal so as the serve all the people who want to listen to the stream at the same time?
- If the app is an aggregator app, such as TuneIn, where will it pull your stream from, will it be direct for your station webstream URL, or through a third party service?
- Who owns the consumption data of your stream? Can you access all available data about each listener? Who else can access data about your listeners?
- Who is monetising your feed? Do they insert their own ads? Do they insert pre-rolls?
- Do you get paid for ads on your stream, or does one of the companies in the chain take all or some of that money?
- Who manages music copyright issues and treaties?
- Can any company in the chain shut off your feed at their whim? Have you read the terms and conditions and know exactly what they can do with your stream and can you contact them quickly if there is a problem?
- If on a smart speaker, has the smart speaker done any deals as to a preferred supplier? If you ask for ‘ABC Radio Sydney,’ will that come to the smart speaker from the ABC Listen App, or the CRA RadioApp, or TuneIn, or iHeartRadio or something else?
- Spoiler alert… the answer to the above question is that Amazon Alexa smart speakers use CRA’s RadioApp as the default choice for a known radio stations name, but if the name changes and the RadioApp doesn’t recognise it, the next choice is TuneIn unless a station has set up something different using Alexa’s new Radio Skills Kit. CRA, under Joan Warner, Kath Brown and Jaime Chaux, worked with the whole industry to develop RadioApp, which shares data equally with all participants. More about the Alexa Radio Skills Kit below.
- If you use a Google Home speaker rather than Alexa you can set it up to give you a preferred feed source, but most people don’t do that, they just accept the default, which is TuneIn. Apple HomePod is the third most used smart speaker type in Australia.
- Will the voice assistant interface understand your Australian accent?
- Will smart speakers recognise their geo-location and prioritise Australian culture and media first, unless specifically requested to deliver a station from another country? At the moment the answer to this is no in my experience, I often request an Australian station, only to be delivered a station from America.
- How will smart speakers interpret your voice command? If I say ‘Play Today FM,’ will I get Hit network’s 2Day 104.1FM Sydney from LiSTNR, or from RadioApp, or Today FM Ireland, or 2dayFM Fiji, or 2Day 95.1 Nigeria… you get the idea.
- If I ask for ‘happy music’ or ‘country music,’ will it pull that feed from a radio station in Australia or from a music streaming service or international station?
- If I ask for ‘Sydney news,’ will it play me ABC702, 2GB, WSFM News, Triple M News…?
One of the new developments now offered by Alexa is the Radio Skills Kit (RSK) that provides radio stations worldwide with a no-code solution to add and manage their stations directly on Alexa. With RSK, stations can create a direct stream for their customers, and access their own analytics, independently of CRA.Both RadioApp and RSK are free for stations to use.
With the RSK dashboard, networks can easily update their streaming URLs, add alternate station names, update logos and artwork, and publish these changes live on Alexa. This allows stations to maintain an optimal experience for customers, including simplified voice commands like “Alexa, play [station name]” and “Alexa, wake me up to [station name]”.
There are a lot of steps in the path from studio to listener if you are consuming your audio on a smart speaker or an app. Smart humans can specify what feed they want on their smart speaker by giving specific instructions such as ‘from iHeart Radio play WSFM,’ but most people don’t do that, they just ask for their choice by station name.
Throughout those steps the possibility of compromised signal quality, audience hijacking, data theft, unauthorised add insertion and other problems is much higher than in the old broadcast transmission path.
But audiences are migrating to digital channels, smart speaker and radio app consumption is growing, so we must use the new pathways. However, the playing field must be level, so regulation and penalties for breaches is necessary.
The government is working through the challenge with industry and has set out a number of principles and questions as it develops a policy. Some of the most important principles are:
Consistent and reliable access – a regulated access provider must respond to and play a specific regulated radio service when requested to do so by the user of a smart speaker, where this is reasonable in the circumstances. |
No cost – a regulated access provider must provide the required level of prominence for a regulated radio service without the imposition of a fee, charge or other form of consideration. |
No alteration – a regulated access provider would be prohibited from altering the content (including advertising) of a regulated radio service when providing prominence for the service. |
Open access pathway – a regulated access provider would be able to determine the source or pathway for users to access a regulated radio service, unless the user requests a particular source. |
Negotiated standards – the framework would enable the relevant parties to determine any technical standards for the integration of regulated radio services and voice activation software… |
These are very good core principles in my opinion.
Smart speakers are currently found in 32% of Australian households and that number is rising. But you would be surprised how many of them are only used for the simplest of things, because most of them are difficult to set up. Requesting radio stations should be simple and the basics should be built in at the default stage, so that when someone unpacks the box, plugs it in and asks for their favourite station, it will be delivered without hassle. More sophisticated users will inevitably do more, but many consumers just want things to give them what they want without fuss.
As the years progress and audiences shift from broadcast to internet delivered radio, there is the very real possibility that companies (often international companies) that control the pipes and end delivery platforms will gain more power than local broadcasters. Early legislative intervention will help the industry retain some of its bargaining power as technology continues to change and be controlled by third party companies with no allegiance to the Australian radio/audio industries.
The consultation paper says: “In contrast to traditional radio sets (which are generally ‘passive’ in terms of the content they are providing), smart speakers actively mediate access to radio services and can facilitate (or alter or block) the delivery of radio content from the provider to the listener.” Australian radio listeners and broadcasters need certainty and reliability going forward, something that the new legislation seeks to implement.
Australia is breaking new ground in this area, because our industry is thinking ahead. Whatever regulations are developed here will influence other countries over time. To date, most regulatory interventions in overseas markets have focused on the prominence of content, rather than services, but recently the UK has legislated separate prominence frameworks for television and radio services, which have informed the Australian legislation.
Well crafted legislation should benefit both the smart speaker and radio companies. The consultation paper says: “radio stations [will] protect and gain incremental listening on smart speakers, which enables commercial entities to earn incremental revenues and all stations to expand their reach; and smart speaker providers [will] gain from having radio present through incremental product sales, the familiarisation of users with voice commands and apps that support the interface, and through the use of data.”
Consumer research in Australia and the UK found that there are many inaccurate results delivered to voice requests for radio stations. According to an audit conducted some time ago by CRA, only 43% of requests for 370 station call IDs via the Google Assistant on a Google Home device returned accurate results. This accuracy has reportedly improved as a result of subsequent commercial arrangements struck between radio broadcasters and the major voice assistant platforms, according to the consultation paper, but “there is no guarantee that this outcome will be maintained if those arrangements expire.”
The government has made a solid case for the development of a regulatory framework to support access to Australian radio services on smart speakers.
One disappointing thing about the consultation paper however, is that is has left out consideration of radio prominence on car radio entertainment screens. This is expected to be tackled in a separate process, but it could have been either rolled into the earlier tv legislation, or into this smart speaker process. The industry will need to keep up pressure on this issue.
The Department of Communications is inviting written comments and submissions on the consultation paper by Monday, 11 November 2024. The last few pages of the paper contain a number of very specific operational questions that the Department would like feedback on. It is well worth reading right to the end to be able to give the kind of feedback required.
The consultation paper and details of how to submit feedback can be found here.
Analysis by Steve Ahern.
You are concerned about getting linked to other countries. What about within Australia? The networking of both the ABC and commercial for example all metro commercial stations are on networks of one in each capital city, but it also occurs in regional areas with lots of MMM and Hit stations as an example. Remember that Australia is 3 hours wide in summer, 2 hours in winter so if you get the wrong state you will end up with the wrong delayed program.
I would expect that local advertisers do not want to be excluded from listeners in their licence areas nor pay for coverage outside of these areas. Regional areas also have their own local news bulletins.
The practice of promoting a radio station's AM or FM frequency only applies to analog radios. For example, Perth commercial station 94.5 is a common frequency in lots of countries overseas. On digital radio, the frequency is totally different and is not used by the listener to select programs. Instead, there is a list of program streams listed alphabetically. In streaming a transmitter is not used at all.
All radio stations have been given unique call signs for their licences by the ACMA. This will substantially resolve the problem. Automated insertion of a call sign in networked programs can be done in the same way as local advertising.
Digital radio both DAB+ and DRM there is additional data which can be added to the transmitted signal to tell the receiver where the same program can be found so that if the signal becomes unusable the receiver will search for alternatives not only on DAB+, DRM but also FM and AM. This means that if the receiver is in a vehicle it can switch to another transmitter from the same broadcaster. It will always select the best quality sound.