Technology update


Voice control in digital signage: more than a technical issue?


Voice-activation is on the cusp of widespread adoption in digital signage, and while there are un- doubtedly technical issues to be resolved, these might be trivial in comparison to the social and political challenges. AV News reports.


With voice control now an in- tegral part of many consumer devices – Samsung reports integration of its Bixby voice technology into fridges and appliances, to smartphones and wearables. Bixby started as a practical way to use your voice to interact with your phone. Now, it is evolving into a scalable, open AI platform to support all of the devices in a consumer’s life. Earlier this year, Samsung introduced the new Bixby – a more conversational, more per- sonal and more useful intelli- gent assistant. Bixby keeps the


providers can access tools to bring Bixby to more people and devices around the world. “Our goal is to offer devel- opers a robust, scalable and open AI platform that makes it easy for them to launch and evolve the amazing experiences they create for our users,” said Kyunghak Hyun, Product Man- ager of the AI Product Manage- ment Group at Samsung. “As Samsung’s ecosystem of de- vices continues to grow, it cre- ates more points of contact not just for Bixby, but for a growing number of third-party services as well. Developers have more


In a conversation with Jeff Hastings, CEO of BrightSign, he


acknowledged that voice


does have a role in digital sig- nage, and that the develop- ment of near field microphones is something that BrightSign is actively pursuing. Another chain of development is the combination of digital signage and the mobile phone where information flow is bidirec- tional. This obviously poses privacy problems in one of the most likely applications to emerge, that of medical advice dispensed from digital signs in doctor’s waiting rooms and


Voice activated digital signage could relieve some of the pressure on the NHS,


dialogue moving forward to get things done. It gets to know you and learns from its users, to personalise how it helps each individual. Bixby serves as a personal assistant, making it quick and easy to complete a variety of tasks from start to finish through a growing num- ber of third-party partners. If a user asks: “Hi, Bixby, give me a restaurant recommenda- tion,” Bixby remembers cui- sine preferences, like Italian, and brings those recommenda- tions to the top of the list. Ask Bixby to buy tickets to a con- cert, and it will ask you follow up questions to find the best seats for the user


Now Samsung is bringing that experience to even more devices and services. To evolve Bixby into Samsung’s intelli- gence platform, the company is laying the groundwork to build a scalable AI platform, making it simpler and easier for developers to create user experiences powered by Bixby. At Samsung Developer Con- ference 2018, Samsung an- nounced that it is opening Bixby Developer Studio, an Integrated Development Envi- ronment (IDE), to developers. It offers access to the same development tools Samsung’s internal developers use to cre- ate Bixby Capsules, which is what developers build to add features to Bixby. It’s all part of Samsung’s unique approach to AI, with the goal of building a scalable, open AI platform where developers and service


P14 AV News September 2019


ways to reach users, and users get more opportunities to make the most of them.”


So how long will it be before Bixby is available for the Sam- sung wall?


Technical barriers


Unlike an application in a con- sumer appliance or a car, where communication


is effectively


one-to-one, voice control of a digital sign has to select one voice command from a cacoph- ony of background chatter. How often have you found the far field microphones installed in voice-controlled


devices acci-


dentally picked up a stray ‘Al- exa’ or ‘Siri’ command from a bystander?


Voice activated digital signage in wayfinding. [photograph Dimension Systems] pharmacies.


Last year Microsoft revealed a push to get OEMs to support Cortana on Windows 10 bet- ter, via support for wake on voice and far-field technology. Far-field technology would al- low users to access Cortana on their PC from up to 4 metres away, much as an Amazon Echo works now, even while music and noise is playing. It requires a PC include a high-quality mi- crophone which meets Micro- soft speech spec 2.0.


Voice on trial


Currently there is something of a battle raging between Micro- soft with its Cortana technology and Amazon with Alexa, Critics


point out that while the two are often seen as synonymous, there are differences. Microsoft has been spreading Cortana around as much as Amazon has Alexa. The difference is that Alexa is a single, cheap device sitting in a room, its relatively easy to implement and permeate the market. Microsoft Cortana is generally considered to be the more powerful potential solu- tion when compared to Alexa, but the product is lacking the direction needed to exploit and perfect Cortana across the wider market for voice control. Amazon’s Alexa, on the oth- er hand, is compatible with a much wider range of devices but both technologies suffer a sim- ilar disadvantage of being pre- dominantly tied to the English language. While Alexa’s prog- ress might be largely attribut- able to marketing success, there are signs that the technology is spreading to more serious appli- cations. For example, the NHS is working with Amazon to allow elderly people, blind people and other patients who cannot easily search for health advice on the internet to access the informa- tion through Alexa. The health service hopes pa- tients asking Alexa for health advice will ease pressure on the NHS, with Amazon’s algorithm using information from the NHS website to provide answers to questions such as: “Alexa, how


pressure on the health service, it has committed to making more services available digi- tally.”


The privacy issue


Placing sensitive medical in- formation on a digital device or network is always a sensi- tive issue. Further privacy con- cerns were raised about Alexa this year after Bloomberg News reported that Amazon employ- ees around the world regularly listen to recordings from the company’s smart speakers as part of the development pro- cess for new services. And Siri, Apple’s voice assis- tant, has long been known to have the power to record private conversations. These audio clips aren’t always just stored on a server. A number of samples are passed along to third-party, human contractors who are paid to listen to them. While Apple trades on the assertion that high-level se- curity comes included with its products’ high prices, it has always been clear that by using Siri, or any voice assis- tant, the user must allow their phone to record and analyse their voice. It’s also known that Google Assistant records and stores audio, but there is at least an option to automat- ically delete your data every couple of months. Amazon’s Alexa stores queries until the user manually deletes them, and both Amazon and Google employ contractors to review a small number of their record- ings. Cortana collects voice data “in order to improve its service”, and Samsung’s Bixby does the same.


Voice is the future?


do I treat a migraine?”; ‘Alexa, what are the symptoms of flu?’; and “Alexa what are the symp- toms of chickenpox?”. The ROI? The Department of Health (DoH) said it would empower patients and hopefully reduce the pres- sure on the NHS by providing reliable information on common illnesses.


The DoH said the NHS would not be giving patients Amazon Echo devices for dispensing NHS-verified information. The Guardian newspaper reports that, by 2020, half of all search- es are expected to be made via voice-assisted technology. “Un- der the NHS long-term plan, which aims to improve the qual- ity of patient care and health outcomes in an effort to relieve


Voice is often described as the future of computing, including voice assistants, voice-rec- ognition technology, ambient computing and the widespread use of smart speakers in the home. But voice is also the future of surveillance: earlier this year the US National Se- curity Agency’s voice-recogni- tion systems, including a proj- ect called Voice RT (“Voice in Real Time”), was reported as having the stated ambition of identifying the ‘voiceprint’ of any living person.


So, for developer of digital signage, the technical issue of developing near-field mi- crophones, the language issue and even the privacy problem might be as nothing when compared to public accept- ability of solutions that listen, record and advise confidential matters!


Continued on page18


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20