One of the key themes throughout my tour of CES 2015 exhibits was the wireless of infrared control of the Smart Home. Using your Smart Phone or Tablet, exhibitors showed how you could easily control many functions in your home.
I am of the belief that inside the home a very natural way to control your Smart Home is your voice. Speaking is a comfortable way to communicate and we do not often do that anymore with e-mail and texting being the means of choice for communication with others. Well, as far as I am concerned, voice is the means I prefer in the Smart Home because it is the most natural means of communications.
There are certainly some challenges with voice such as background noise, loudness and clarity of our words, even slight accents in speech. Conexant demonstrated to me how they have refined the art of voice-control with some innovative, engineering solutions. I was impressed.

To accomplish background noise elimination in a typical home environment, Conexant uses their unique far-field voice processing technology. Far-field voice processing for smart home applications is not based on traditional beamforming. Traditional beamforming requires users and noises to be at pre-defined angles relative to the product’s microphones, which are not practical for most smart home applications. On the other hand, the Conexant solution takes real-world environments into consideration and is based on Blind Source Separation (BSS). BSS enables excellent speech recognition performance from up to five meters away, at any angle, and with only two microphones – even when noise is in the same direction as the voice.

This Linux-based solution is the newest addition to a portfolio that already includes Conexant’s software solution for Windows and Android operating systems, Conexant’s DSP CODEC solution for embedded applications, and its high dynamic range microphone ADCs for optimal far-field pick-up.
This unique combination of hardware and software is a winning design that has enabled true and accurate voice control in the Smart Home.
What are your thoughts regarding Voice, Wireless or Infrared Control?
Voice recognition during driving on vehicle always challenges vehicle engineer regarding on how to remove the road noise and engine noise, as well as noise when other vehicle is passing. Noise and echo cancellation has a limited capability. It would be expected that audio smart might solve these issues.
It is good time we can say that we can talk to our home, although to me it sounds like some sci-fi movie or hi-tech lab. Voice regocnition has always been tough part either in unified communication or in Mobile phones.
Voice Control seems to be the most intutive way of controlling anything. The user don't need any thing to carry and doesn't need any special knowledge to operate and control the things apart from basic command words.
The very big challange is the Voice Recognition itself. It needs lot of processing power unlike other interfaces like wireless and IR. Apart from noice calcellation, understanding correct word has been the biggest challange to solve I think. I do not very good experiece with google voice search in my android phone. The main reason I think is my Indian accents. The same could happen with any voice recognition system which is being used around the world.
@Steve, thank you for this interesting article, I think that the Blind Source Separation is really promising, especially considering the presence on the market of smart microphones ICs, based on MEMS.
I think that the wireless control is the key for all modern smart home appliances, and this control might be bidirectional and there is room for many types of functionalities: remote assistance of a patient , safety purposes, smart lights, etc…
> I do not very good experiece with google voice search in my android phone. The main reason I think is my Indian accents. >>>>>>>>>>
For me also the same kind of experience. I also does not find it enjoyable voice commanding in my phone. That is a challnge that needs to be looked at.
>>>>>>>>>>>> Blind Source Separation is really promising, >>>>>>
yes it is indeed. This will open up many opportunities and encouragements to blind people specially in their day today life.
@ue2014: you're absolutely correct! Some useful applications for make easier the life of blind people and moreover for patients that might be followed by remote from a medical center, there's a lot of potential audience.
@ue2014: I add that many interesting applications for blind people have recently appeared , like for example the Arianna Navigation system for smartphones, a very promising solution. Try to google it
I am little curious to understand what if someone wants to impose low complexity constraint on the signal,i.e. if one requires not the whole signal.
“I do not very good experiece with google voice search in my android phone. The main reason I think is my Indian accents.”
@Vishal: You can change the voice accents and customize it to Indian English accent. That should act more semaless in your case.
“The very big challange is the Voice Recognition itself.”
@Vishal: Yes its very challenging and complex. Now a days with race for the touchless technolgy for smart phones lots of companies are working on different applications.
One of them is Motorola, they have comeup with the voice commands based touchless tech to operate the Moto-X series. This works nice even in the presence of noise environments once the device has been configured/trained.
@Steve,
This is really a intuitive way of control and the Blind Source Separation (BSS) as i read from the blog and comments looks promising.
It will be fascinating to know if it works when watching a movie.
If I am watching a movie with high decibels and there is a dialogue, “turn-on the lamps”. Does recognize this system recognize this and issue a command, or is this treated as some kind of noise and ignore it?
Oh I didn't know that is possible in Android. I must give it a try.
Is it a google account setting or Andoid OS setting?
samicksha: I guess the processor requires only the part of signal that is enough to separate the different voice sources.
The “Moto-X series” is very interesting to me, the voice communication is an effective mean to control smart objects through a smartphone, with installed dedicated appliances to drive the objects. This solution is compatible for example with smart bluetooth technology.
speech recognition is a very complex problem, evaluated in terms of accuracy and speed and here accuracy still seems to be asking for more attention.
Very true DaeJ, It will be a very difficult thing to get the exact command while having a noises around the vehicle. But for the home this will be not a big issue.
Yes Vishal you are highlighting a very valid point there. I feel that there will be a high accuracy by having combination of a physical action with a voice command and having a predefined set of command.
Indeed etnapowers wireless control is the future for sure. There will not be an any demand for the wired controller in the market.
“Is it a google account setting or Andoid OS setting?”
@Vishal : It is an Andriod setting. Look for the voice search settings and Choose English (India). You will hear back a response sounding like an Indian Female voice.
Thanks.
@etnapowers: I am not able to relate the the smart bluetooth technology and the voice-comm on smart phone. In particular the Moto-X has a always on microphone (low-power) to listen for any voice commands and action thereof.
I am missing something to build a relation there, Please throw some light.
Thanks.
“speech recognition is a very complex problem, evaluated in terms of accuracy and speed and here accuracy still seems to be asking for more attention. “
@samicksha: Agreed, I think that the modern ICs based on a smart designing approach, like the smart microphones and the high bit-rate microprocessors, might overcome the challenges of the speech recognition.
@Ranasinghe: You're absolutely right, the wired control has been utilized massively in the past, the wireless is the future, but , like any new solution, it presents some challenges, for example the bandwidth required for big volumes of data to be transmitted through the communication network.
@amrutah: the smart bluetooth technology has the potential to recognize the presence of a smartphone in a controlled environment, the smartphone has the speech recognition application, thus the two things may be combined to achieve a smart system with many functionalities.
” I also does not find it enjoyable voice commanding in my phone. That is a challnge that needs to be looked at. “
True ue2014, Sometimes it is wasting of time when we have all the command at our finger tips.
“Some useful applications for make easier the life of blind people and moreover for patients that might be followed by remote from a medical center,”
Etnapowers, those are thing that will highlight the value of human being.
” I guess the processor requires only the part of signal that is enough to separate the different voice sources”
Etnapowers, I believe that it will depend user to user, and also it should be pre programmed according to the user.
Etnapowers, it I a great thing which is integrating the Bluetooth and the voice recognition at the same time cause that it will help to minimize the capturing of the other noises.
>>>>> it presents some challenges, for example the bandwidth required for big volumes of data to be transmitted through the communication network. >>>>
I am sure future will find answers for all these challenges and this concept would be our future. It has many uses to go ahead with Wireless options which could not be ignored.
>>>>>>>>>>> “Some useful applications for make easier the life of blind people and moreover for patients that might be followed by remote from a medical center,”
Etnapowers, those are thing that will highlight the value of human being.>>>>>>
True and correct. Above are very usefull. But security and the accuracy of such implimentations needs to be looked at very carefully as these involves with Human Lives directly.
“I am not able to relate the the smart bluetooth technology and the voice-comm on smart phone. In particular the Moto-X has a always on microphone (low-power) to listen for any voice commands and action thereof.”
@amrutah: You're right that there are voice-based commands you can use on smartphones. However, they're only there to activate smartphone functions. What the blog is talking about is how those commands (used via your phone) can be used to control features of your smart homes. This itself is much more complicated than that.
@etnapowers: I'm not sure if Bluetooth alone can be used as the sole form of wireless networking. I think it is only for a short range and may not work under all circumstances. You may need a combination of bluetooth and WiFi for example to cover a wider area.
” However, they're only there to activate smartphone functions”
@Tzubair: I agree that there are certain commands that are needed to activate the phone but later you can input your command like calling or texting a particular person, voice searching the web or phone for certain files or videos.
As etnapowers suggested, this can be extended by using the smartphone with bluetooth which can further increase the connectivity between the smart home devices and the voice enabled smartphone.
“You may need a combination of bluetooth and WiFi for example to cover a wider area.”
@Tzubair: I think there are already smart home solutions which you can control using the phone texting, Home communication networks (ethernet and wifi).
As suggested in the blog this can further be enhanced if we can talk to the device. I agree that a combination of the communication systems can make them reliable and robust (although that can increase the cost).
@amrutah: Home automation solutions are really approaching faster than we expected, only to be hindered by the assortment of standardizations framed on the components of a home automation system. Another factor is how the data connects to the cloud, now that is a place where IOT and smart home developers are really worried about.
@tzubair: judging from a voice recongnition perspective, this is quite possible, not only just smartphones but most high end cars now use VR systems as an exra precaution to that the car is not stolen. Similarly if we're about to live in a smart home, it needs VR systems.
“this can be extended by using the smartphone with bluetooth which can further increase the connectivity between the smart home devices and the voice enabled smartphone.”
@amrutah: This can only happen when there's a network involved. A smartphone in isolation may not be able to achieve this alone. You'd need a smart network which connects all the devices together to a network and allows your smart phone to have control over the network. This may also involve IoT at some point.
“this concept would be our future”
True ue2014 even today we are used to this tech and we are depending on it in many times. As you says this will be our future
@ranasinghe: Yes and mainly because that it provides what we want in a very simpler but more effective manner than it used to be in the earlier days
@ue2014: Yes you can go ahead with wireless technology but there are some security issues in it which needs attention these days. If it can be solved then surely it's a very good option to have
Tzubair, it is a great idea to have a integrated system with both Bluetooth an WIFI, But I feels that the security will be low when it comes to WIFI.
@dassa.an: Well I don't think so because there are security features in-built with the system itself
@chri: Well I guess you are referring to passwords here. True but its easy to break those and get in to the network.
amrutah, I believe that the heard thing is to develop the voice controlling mechanisms rather than the communications.
dassa.an exactly simplicity is very essential when it goes to the market, other vice existing system will more reliable than the new tech.
@ranasinghe: Yes you do have to make things look more simple. Design is what matters
@ranasinghe: Well in a way yes but those two are two different things.
@dassa.an: Not if you can put a complex password. That's not difficult as long as you can remember it
“But I feels that the security will be low when it comes to WIFI”
@Ranasinghe: I think WiFi would be much secure if you have WPS based encryption running on the network. Even if you think it isn't, there isn't a great deal of secure data that will travel over the network.
@dassa.an: Not if you can put a complex password. That's not difficult as long as you can remember it
Maybe store it in a word document and inside the cloud so that you don't forget the password and lose the password even if your machine breaks.
@ranasinghe: Yes you do have to make things look more simple. Design is what matters
Too simple designs aren't considered to be stimulating for the consumer to invest in. As long as there is some bling (or flashy parts in a design) one can think that only the reputation of the company would make someone buy a really unattractive piece.
@amrutah: This can only happen when there's a network involved. A smartphone in isolation may not be able to achieve this alone. You'd need a smart network which connects all the devices together to a network and allows your smart phone to have control over the network. This may also involve IoT at some point.
@tzubair: I agree with you. A smart phone alone cannot hold every smart device in its place no matter how powerful the smartphone is. You need a wireless gateway device.
@amrutah: Home automation solutions are really approaching faster than we expected, only to be hindered by the assortment of standardizations framed on the components of a home automation system. Another factor is how the data connects to the cloud, now that is a place where IOT and smart home developers are really worried about.
The “Smart Home” provides intelligent networking of devices and systems by information and telecommunication technologies and opens completely new possibilities of living and working.
In the Smart Home Energy management, consumer electronics, comfort, safety and health are more cost-effective or convenient. Norms and standards provide the basis for the technical integration and support the development.
Through the German Standardization Roadmap “Smart Home + Building” the VDE | DKE German Commission for Electrical, Electronic & Information Technologies of DIN and VDE advanced a great step in this area.
The Standardization Roadmap provides an overview of the current status and trends, the valid norms and standards and which are to be developed now. At the same time it is an essential ingredient for the international market development in this field.
Another factor is how the data connects to the cloud…
@SachinEE: This is one of the important factor that is necessary when we need to control home appliances through the mobile or bluetooth or voice communication. What is the hindering factor?
It would be expected that audio smart might solve these issues.
@DaeJ, I totally agree with you. I think what makes this product unique is Blind source seperation (BSS) which makes it more useful.
I think that the Blind Source Separation is really promising, especially considering the presence on the market of smart microphones ICs, based on MEMS.
@etnapowers, I totally agree with you. I am curious to know if we can use the same technology in other fields as well especially where we require noise seperation.
I add that many interesting applications for blind people have recently appeared , like for example the Arianna Navigation system for smartphones, a very promising solution. Try to google it
@etnapowers, thanks for sharing this info. What makes Arianna useful is that it is a flexible and low cost navigation which makes it more affordable.
Not if you can put a complex password. That's not difficult as long as you can remember it
@chirshadblog, well complex passwords are always difficult to remember. I think its always advisable to use backup systems to save such passwrods.
@SachinEE: I agree, the voice recognition is the way forward with the systems around. This is mainly helpful for people with vision and physical disability. The VR systems are on the way of maturing is the way going ahead.
@tzubair: I disagree to this. If the smart systems are bluetooth enabled then we can connect to the smart systems using the voice enabled smartphone. Although additional networks can provide the robustness needed (like error corrections, cloud analytics and storage).
You'd need a smart network which connects all the devices together to a network and allows your smart phone to have control over the network. This may also involve IoT at some point.
@tzubair, I totally agree with you. I think we have to use IoT if we cant to connect all the devices together. But I think only conern is security.
If the smart systems are bluetooth enabled then we can connect to the smart systems using the voice enabled smartphone.
@amrutah, I think only disadvantage of bluetooth is the range. Bluetooth range is very less and am not sure if we can increase that range.
You may need a combination of bluetooth and WiFi for example to cover a wider area.
@tzubair, I agree with you. I am not sure why bluetooth range is so small not sure if its feasible to increase that range.
In particular the Moto-X has a always on microphone (low-power) to listen for any voice commands and action thereof.
@amrutah, similar features are available in x-box which is in sleep mode most of the time but is still listening to the voice commands.
@SunitaT0: I think the bluetooth was meant for short range communications and increase the range will call for more power which overcomes its sole purpose.
I think the bluetooth was meant for short range communications and increase the range will call for more power which overcomes its sole purpose.
@amrutah, so are you saying that increasing the power increased the range ? If we have some application where we can afford to spend more power we can have larger range ?
@SunitaT0: The increase in range will call for more power and extra communication channels needed in the same frequency band. The more important issue with the bluetooth I see is the security. How many layers of security is added to the communication frame as compared to the wired or wireless networks.
You can change the voice accents and customize it to Indian English accent.
@amrutah, thanks for shaing this tip. I didnt knew it was possible to change the accents as well.
I am little curious to understand what if someone wants to impose low complexity constraint on the signal,i.e. if one requires not the whole signal.
@samicksha, what exactly you mean by constraint on signal ?
If I am watching a movie with high decibels and there is a dialogue, “turn-on the lamps”. Does recognize this system recognize this and issue a command, or is this treated as some kind of noise and ignore it?
@amrutah, that is a very good question. I would also be curious to know if the system identifies that as command.
@Ranasinghe: The human being is the center reference for all the future applications of the electronics systems, thus I think that the contribution of electronics to the quality of the life of all of the people will be a great contribute in the near future.
@Ranasinghe: you're talking about the voice recognition of a particular user , and the memorization of the related data, correct?
Agreed. It's one of the opportunities of the bluetooth technology, I guess that the smart bluetooth solution will be widely adopted in multiple typologies of future applications.
I think there are many interesting scenarios open to the noise separation purpose.