|Europe's leading residential technology trade magazine||Subscribe|
Technology: More than Touch - Voice and Gesture Control of Home Automation (6/6/2012)
By Khalifa A. AlJaziri, e-Home AUTOMATION
Traditional solutions for automation control have largely focused on extensions to the humble light switch. We then moved on to the infrared or radio frequency remote control, and eventually arrived at schedule-based control options that switch lights on and off, open and close curtains and perform other tasks at set times of day.
While there have been some more intuitive control options that didn't really catch on (who remembers hand clapping to switch on lights?), ultimately, what could be more natural than speaking or gesticulating a command? Indeed a number of companies, including our own, already offer systems for voice control that have a high degree of accuracy, and with the rise in popularity of technologies such as Apple's Siri and Microsoft's Kinect, voice and gesture control are now realistic options for automation companies to consider. These technologies may have a certain 'gimmick' factor, but they are potentially very useful, especially in assisted living applications.
Voice Recognition Considerations
At present, voice control is widely available and is most easily implemented using PCs with Microsoft operating systems, such as Vista and Windows 7. The built-in voice recognition software of these operating systems allows home control via software and a microphone, usually with no 'training' of the system required.
People are also hacking the Apple Siri (Speech Interpretation and Recognition Interface) to make control of a variety of devices possible via an iPhone. Siri sends your voice command or query to the Apple Siri servers which then send back the appropriate data. Hacking involves mimicking the interface between Siri and the Apple servers in order to make it work with other systems, but this, of course, is not recommended for professional applications.
When designing our voice control system, we decided that it was important to give the system a name. This key word would be listened for by the system, in order to know that a command was being issued. For those of you familiar with the TV show Star Trek, this is the same principle of prefacing a command with the word Computer, e.g. 'Computer, make me a cup of tea!' We typically give the system a human name, which falls into the idea of it being a 'digital butler'.
Another important consideration is user feedback - if the voice recognition system is only one-way, it can be frustrating or confusing to the user to operate. To make for a more satisfying and cohesive user experience, we decided to have an 'announcer' module that talks to the user via in-ceiling speakers. For this, the Microsoft text-to-speech engine is employed, and we use a high-quality voice module, such as AT&T Natural Voices for English language, and Acapela for others.
This requires a little bit more programming time for the integrator than normal, as each voice command must be programmed (along with any obvious variations). The response of the system must also be coded and linked to the home automation control hardware. For example, 'Computer, turn all living room lights on!' must be linked to a programmed macro that affects all of the appropriate devices in that room.
In terms of microphones, there are several options. Obviously, you don't want to have to wear a microphone around the property, so good solutions include the in-ceiling array microphones that are typically used for conferences, and devices that can be picked up and used when required - our favourite is also a Microsoft Media Center remote control, with the microphone built in and activated by tilting.
When it comes to system installation design, there are several considerations: in how many rooms is voice control required? What are the distances that must be covered - does the client require control in one room - typically the living room or bedroom - or in several rooms? How many PCs are then required for individual room control?
The great news is that voice control for homes is a rock-solid and off-the-shelf technology that is available today - despite most people being unaware of it and thinking of it as 'something from the future'.
Gesture control for home automation is in its infancy. Although researchers at Microsoft have already commented on the potential of the Windows 8 version of Kinect (which differs from the Xbox version) for use in home automation applications, it is not yet clear how this would work in practice. As home automation system developers, we want to provide technologies that people will use, and we also spend a lot of time thinking about how people will use them. Outside of a gaming environment, there isn't much to draw on in terms of how people will react to devices such as Kinect, so we are waiting for the release of Windows 8 later in the year to see how much demand there is for this type of technology in a different setting.
Whilst devices such as Kinect are clearly very impressive pieces of technology, the infrared laser system and CMOS unit that they employ is not the only approach to gesture recognition. An alternative technology is also being investigated that utilises naturally-occurring ambient electromagnetic radiation: the human body produces a small 'signal' as it interacts with this ambient electrical field, and the systems being developed use algorithms to interpret and harness that interaction.
Voice and gesture control offer interesting alternatives to integrators and end-users for extending their control options in a home automation system. Beyond the wow and gimmick factors, the technology offers tangible benefits to disabled and elderly users.
Systems can be constructed using off-the-shelf hardware and software, and a solution can be deployed for as little as a few hundred dollars (excluding the home automation control hardware). However, the costs are still high in terms of mass-rollout for assisted living systems.
Although there is no doubt that these technologies are exciting, we believe that the future holds some fascinating and extraordinary technology advances in this and other areas. For example, the continuing work into nanotechnology and artificial intelligence is making progress, and anyone familiar with Ray Kurzweil will know that he believes a technological 'singularity' is just years away - perhaps we could one day see control using devices we wear in our clothing, and perhaps our 'digital butlers' will be advanced artificial lifeforms!
However the technology progresses, alternative methods of home control are the future, as we are just beginning to see with the technologies that are available today.
Khalifa A. AlJaziri is the Group Managing Director of e-Home AUTOMATION LLC, a global specialist in automation solutions and energy management, offering integrated solutions across multiple market segments including construction, real estate, facilities management, security, government, telco and energy sectors.
| newsfeeds | subscribe
to newsletter | submit
a link |