Category: Voice-driven home assistant platforms

Article

The mute button on your Echo or other smart speaker is important if you want greater control over the voice assistant

You Should Mute Your Smart Speaker’s Mic More Often | Lifehacker Australia

My Comments

An issue that will plague people who own smart speakers like the Amazon Echo or Google Home is the device interjecting with responses when you or someone else unintentionally say certain words.

This is because these devices are typically set up to listen all the time for a particular “wake word” that actually invokes the voice assistant. It is part of the machine-learning that drives these voice-assistant platforms to understand what you say.

But you can have some control over these smart speakers to avoid this behaviour. Each smart speaker or similar device will have a hardware mic-mute switch on them, highlighted with a microphone icon on that switch. This effectively turns the device’s microphones on or off as you need and the article recommended that if the device is too “hair-trigger”, you should enable this function unless you are actually intending to interact with the voice assistant. This procedure would be similar to how you would work a voice-driven personal assistant that is part of your smartphone’s operating system where you deliberately press a button on the device or a Bluetooth headset to invoke that assistant before you say the wake word.

Beware of the situation where the button will light up when you enable microphone muting on your smart speaker. Here this may be a point of confusion because some users may think that the device is “ready” to speak to when it is infact not able to take commands. You may have to be familiar with how your smart speaker looks when it is ready to accept commands, including any lights that are on.

If your voice-assistant platform has the function to “edit” what has been captured like what Amazon Alexa can do, this function can be used to fine-tune what the voice-assistant is meant to respond to. The same control app or Website can be used to manage your smart speaker’s microphone sensitivity or change the “wake command” that you say when you start interacting with the assistant.

They even recommend that you disconnect the smart-speaker device from the power if you don’t intend to use them. Privacy advocates even suggest doing that in areas where you value your privacy like your bedroom or bathroom.

But personally I would at least recommend that you are familiar with the hardware controls that exist on your smart speaker or similar device so you are able to have greater control over that platform.

Article

The BBC is working on a voice-driven assistant of its own

BBC planning Alexa rival | Advanced Television

BBC’s Beeb voice assistant goes beta | Advanced Television

BBC launches voice assistant that will learn regional accents | Mashable

From the horse’s mouth

British Broadcasting Corporation

What’s next for Beeb: The new voice assistant from the BBC (Press Release)

My Comments

The BBC have started work on the first voice-driven assistant platform that is designed in the UK and for British conditions. This platform is named Beeb which is one of the common affectionate nicknames for that public-service broadcaster.

It will be initially rolled out to UK participants of Microsoft’s Windows Insider program and is at beta stage, thus limiting it to use on regular computers at the moment. It is based on the BBC’s iPlayer content-directory effort and will be primarily about bringing up audio content like live and on-demand radio content, music or podcasts. It will also be able to bring forth local news and weather information.

… optimised for UK accents and likely to place the UK in the voice assistant sphere

There will even be the ability to respond from material written by BBC comedy writers when you ask for a joke, including the ability to ask for a joke from a particular show. This is thanks to the amount of intellectual property that they have built up over the years including all those legendary sitcoms and other comedy associated with British telly.

Beeb will be voiced with a male voice that has a northern English accent. As well, its initial setup will have the user determine what UK regional accent they have so it understands their accent better. This is part of an effort by the BBC to preserve local British accents in the face of other voice-driven assistants which force users to use standard British English pronunciation over regional accents.

There is an intent to roll it out to other devices which are software-programmable and it will also to be part of revisions to the iPlayer app and BBC Website. But personally, I see the BBC’s Beeb effort as a candidate for the Voice Interoperability Initiative driven by Amazon and Microsoft that allows a device to run with multiple voice assistants and respond to each assistant’s “wake word”. That is an activity that the BBC are infact a participant in.

Support for regional and local accents

But personally, I see the Beeb voice assistant as opening a path for UK companies to develop their own voice-driven assistants that respect UK language, dialects, accents and culture. It can also be an example of fine-tuning a voice-driven assistant platform to work with the various local accents and dialects used within a country or language and avoid steering the user towards what is seen as a standardised pronunciation for that language.

What will eventually need to happen would be to allow automatic detection of a user’s accent and to work with that accent automatically when they speak, rather than requiring a user to determine which accent they are using during setup. Having to determine which accent you are using during the setup phase can be a problem for households with members that come from different regions.

I would also see efforts like Beeb end up being about local-speech “modules” for voice-driven assistant platforms that enable these platforms to support a country’s or language’s local peculiarities when it comes to regular use. It will then avoid the need for people to resort to using “standard” diction rather than the accent they are comfortable with to deal with voice assistants. Similarly, it could be about different “voicings” that maintain local characteristics for the assistant’s speech.

Who knows what this could mean for making voice-driven assistants locally relevant?

Smart speakers like the Amazon Echo are outselling traditional radios

The traditional radio broadcast industry are finding that the smart speaker as a threat to their business models.

This is because that there are more Amazon Echo, Google Home or similar smart speakers being bought than traditional radio sets. It is in addition to us using smartphones that don’t have traditional broadcast-radio tuners as our “go-to” information and entertainment devices.

Although these smart speakers can, at your voice command, pull up a traditional radio station thanks to TuneIn or similar Internet-radio directories, an increasing number of users are using them to summon podcasts or music playlists through the various podcast and music-on-demand services.

Pure Sensia 200D Connect Internet radio – an example of how to keep the traditional radio relevant

At the moment, traditional radio whether through traditional broadcast technology or Internet streaming is primarily being listened to in the car or at businesses we frequent. It is also being seen, whether for information or entertainment, as a valid casual-listening content-source by Generation X (people born from the late 60s to the early 80s) and prior generations thanks to it being seen that way for a long time. This is due to the ubiquity of increasingly-affordable radio sets in many different form factors along with radio stations making a strong effort to keep listeners tuned to their output.

It is although advertisers and others have seen and are seeing the younger generations as “where the money is”. Here, they end up sponsoring podcasts or playlists to reach that audience with their message in order to stay relevant.

The ABC, like other traditional broadcasters, are offering their own podcasts, whether to do with an existing radio show or not

But what can be or is being done about this? At the moment, traditional radio stations are creating podcasts, whether as a byproduct of an existing radio show or as a new product. Similarly as I have experienced, most radio stations are planting their regular broadcast output on the Internet and making sure this still happens so as to work with smartphones and smart speakers. It is even though they face battles with music rightsholders and sporting leagues about international streaming rights for music or sports content.

RadioDNS “hybrid radio” has surfaced as a way to bring together traditional radio and the Internet. The key method offered by this platform is through a “single-dial” approach that provides a seamless handover between local radio frequencies / DAB multiplex locations and Internet streams for the same radio station.

This Internet radio is tuned in to Heart London and is playing the same audio as what would be delivered on FM or DAB from the “Turn Up The Feel Good” station within the London area

Reliance on Internet audio streams as often done with smart speakers and smartphones can be problemsome if you don’t have the right kind of network and Internet connection. This represents the typical home or small-business network connected behind most home / small-business routers.

You will run in to problems with setting up a smart speaker or similar device to work with a headline public-access / guest-access Wi-Fi network that depends on Web-based authentication or having these devices work with an enterprise-grade network that uses per-device-based authentication approaches. It also includes dealing with mobile broadband services that charge an arm and a leg for continual bandwidth use but services that operate in a highly-competitive market may make this factor easier.

The stations listed on the TuneIn Internet radio app are the Internet-hosted simulcast stream of their regular radio output

Similarly broadcast-radio technology tends to appeal to listenership on battery-operated devices because the technology associated with it is optimised to work for battery efficiency. It is due to the broadcast-radio technology working on a one-way approach to receiving the radio signals rather than being dependent on a two-way transceiver demanded of Wi-Fi or mobile-broadband.

What can be done to bridge these technologies

One approach would be to have an Internet radio that also receives radio content via broadcast technologies work with at least one of the common voice-driven home assistant platforms.

This can be in the form of the radio working alongside a smart speaker based on the common platforms and using RadioDNS to pull up local radio stations under voice control.

An Internet radio can also serve as a speaker for online audio resources like on-demand music services, podcasts and Internet radio especially if the radio doesn’t have network-audio / Internet-radio functionality. The latter concept is being underscored with the Google Assistant platform where you can direct audio from an online-audio service to a device that supports the Google Chromecast protocol. Even if the radio has network-audio / Internet-radio functionality, it could be part of a voice-driven home-assistant platform, which a lot of manufacturers are heading towards and can be of relevance for the “big sets” like hi-fi systems and the network multiroom audio platforms.

A cheaper option could implement RadioDNS across a Bluetooth link with the voice-driven home assistant platform handling the RadioDNS logic. It may require the creation of a Bluetooth profile for sending RadioDNS-specific data between the radio and the smart speaker’s platform i.e. a set-appropriate pointer to the station on the broadcast bands.

It can also be about an Internet-radio / smart-speaker combination device, like the many combination devices available over the years that integrated radio reception and at least one other function. Such a set would have the ability to be an Internet radio but it would have a microphone array and a button to activate Amazon Alexa or Google Assistant, whereupon you would have the full “smart speaker” abilities of an Amazon Echo or Google Home speaker. As well, it would tie in with the RadioDNS functionality to pull up stations on the local wavebands as if you are pulling them up using the assistant’s Internet-radio functionality.

Conclusion

To keep the classic radio medium going, the manufacturers, broadcasters and other stakeholders need to look at whatever technologies can be used to make it relevant in this day and age.

Articles

The Google Nest Mini smart speaker – a follow on from the Home Mini smart speaker and having its own local processing to improve Google Assistant’s responsiveness

Google Nest Mini gets louder and gains onboard Assistant processing | SlashGear

Google debuts Nest Mini with wall mount and dedicated ML chip | VentureBeat

From the horse’s mouth

Google

Nest Mini

Nest Mini brings twice the bass and an upgraded Assistant (Product Blog Post)

My Comments

The Google Nest Mini smart speaker, which is the successor to the Google Home Mini smart speaker, shows up a significant number of improvements including a richer sound. But it has also come about with the idea of locally processing your voice commands for better Google Assistant performance.

The traditional approach to processing voice commands that are said to a smart speaker or similar device is for that device to send them out as a voice recording to the cloud servers that are part of the voice-driven assistant platform. These servers then implement their artificial-intelligence and machine-learning technology to strip background noise, interpret the commands and supply the appropriate replies or actions back to that device.

But Google has improved on this by using a leaf out of the book associated with edge-computing technology. This is where some of the data storage or processing is performed local to the user before the data is sent to a cloud computing system. Here, Google uses a dedicated machine-language processor chip in their Nest Mini smart speaker to do some of the command processing before sending data about the user’s command to the Google Assistant cloud system.

It reduces the idea of your Google Nest Mini smart speaker being a simple conduit between your home network and the Google Assistant cloud. The key benefit is that you see a quicker response from the Google Assistant via that device. You also have the benefit of reducing the Internet bandwidth associated with handling the voice-driven home assistant activity, avoiding reduced performance for online gaming or multimedia streaming.

Google is working on taking this further with having Google-Assistant-based devices that have this kind of local processing process logic associated with user requests and programmable actions locally. It also includes keeping the logic associated with the Assistant liaising with other smart devices local to your home network, allowing for improvements to performance, user privacy and data security.

It could be seen by Amazon and others as a way to improve the performance of their voice-driven home-assistant platforms. This is more so where the competition between these platforms becomes more keen. As well, there could be a chance for third-party Google Assistant (Home) implementations to look towards using local processing to improve the Assistant’s response.

An issue that will crop up is having multiple devices that have this kind of local processing existing on the same home network help each other to increase the voice-driven assistant’s performance. This can also include using a software approach to make the devices equipped with the local processing provide improved performance for those that don’t have this processing. It will be an issue with the likes of Google Nest Mini and similar entry-level devices that appeal to the idea of having many installed around the house, along with the idea of equipping smart displays with this kind of local processing.

What I see of this is that the use of edge-computing technology is coming to the fore as far as improving responsiveness in the common voice-driven home-assistant platforms.

Article – From the horse’s mouth

Google Nest

How ultrasound sensing makes Nest displays more accessible {The Keyword blog post)

My Comments

Google is implementing in their Nest Hub smart-display products an automatic display-optimisation technology that is based on technology that has been used for a very long time.

Ultrasonic technology has been used in various ways by nature and humans to measure distance. In nature, bats and dolphins which don’t have good vision use this approach to “see” their way. It is used extensively in military and civillian marine applications to see what is underneath a boat or around a submarine and is also used as a common medical-imaging technique.

As well, in the late 1970s, Polaroid implemented ultrasound as part of their active autofocus system, which ended up as a feature for their value-priced and premium instant-picture cameras. Here, this was used to measure the distance between the camera and the subject in order to adjust the lens for proper focus. There were limitations associated with the technology like not being able to work when you photograph through a window due to the ultrasonic waves not passing through the glass.

But Google has implemented this technology as a way to adjust the display on their Nest Hub smart displays for distant or close operation. The front of a Google Nest Hub has an ultrasonic sensor that works in a similar way to what was used in a Polaroid auto-focus instant-picture camera.

But rather than the Polaroid setup being about using the distance measurement from the ultrasonic sensor to adjust a camera’s lens, this application adjusts the display according to the user’s distance from the Nest Hub. If you are distant from the Nest Hub, you would see reduced information but the key details appear in a larger typeface. Then if you come closer to the Nest Hub, you would see more detail but at a smaller typeface.

The Google blog article described this as being suitable for older users and those of us who have limited vision. The fact that you have the ability to see key information in a large typeface at a distance can make the Nest Hub accessible to this user group. But others can’t see deeper information unless they are very close to the device.

End-user privacy is still assured thanks to the use of a low-resolution distance-measurement technology whose results are kept within the device. As well, there is a menu option within the Google Home app’s Device Settings page to enable or disable the feature.

At the moment, it is initially being used for timer and current-time display as well as displaying travel time and traffic conditions for a planned journey that you set up with Google Maps. But Google and other software developers who develop for the Google Home ecosystem will add distance-sensitive display functionality to more applications like appointments and alerts.

Some people could see this technology not just for optimising the readout on a smart display but could even be used to ascertain whether people are actually using these devices. This could then be used for such functionality like energy-saving behaviour where the display turns off if no-one’s near it.

But what Google has to do is to license out this technology to allow others to implement it it to other fixed-display-based devices. Here, it could become of more use to many who don’t go for a Google Nest Hub.

but to add more functionality like appointments, alerts, reminders

Articles

Devices like Amazon Echo could support multiple voice assistants

Amazon Creates A Huge Alliance To Demand Voice Assistant Compatibility | The Verge

Amazon launches Voice Interoperability Initiative — without Google, Apple or Samsung | ZDNet

Amazon enlists 30 companies to improve how voice assistants work together | Engadget

From the horse’s mouth

Amazon

Voice Interoperability Initiative (Product Page)

Amazon and Leading Technology Companies Announce the Voice Interoperability Initiative (Press Release)

My Comments

Amazon have instigated the Voice Interoperability Initiative which, at the moment, allows a hardware or software device to work with multiple compatible voice-driven AI assistants. It also includes the ability for someone to develop a voice-driven assistant platform that can serve a niche yet have it run on commonly-available smart-speaker hardware alongside a broad-based voice-driven assistant platform.

Freebox Delta as an example of a European voice-driven home assistant that could support multiple voice assistant platforms

An example they called out was to run the Salesforce Einstein voice-driven assistant that works with Salesforce’s customer-relationship-management software on the Amazon Echo smart speaker alongside the Alexa voice assistant. Similarly, a person who lives in France and is taking advantage of the highly-competitive telecommunications and Internet landscape there by buying the Freebox Delta smart speaker / router and have it use Free.fr’s voice assistant platform or Amazon Alexa on that same device.

Microsoft, BMW, Free.fr, Baidu, Bose, Harman and Sony are behind this initiative while Google, Apple and Samsung are definitely absent. This is most likely because Google, Apple and Samsung have their own broad-based voice-driven assistant platforms that are part of their hardware or operating-system platforms with Apple placing more emphasis on vertically-integrating some of their products. It is although Samsung’s Android phones are set up to be able to work with their Bixby voice assistant or Google’s Assistant service.

Intel and Qualcomm are also behind this effort by offering silicon that provides the power to effectively understand the different wake words and direct a session’s focus towards a particular voice assistant.

The same hardware device or software gateway can recognise assistant-specific wake words and react to them on a session-specific basis. There will be the ability to assure customer privacy through measures like encrypted tunnelling for each assistant session along with an effort to be power-efficient which is important for battery-operated devices.

Personally I see this as an ability for companies to place emphasis on niche voice-assistant platforms like what Salesforce is doing with their Einstein product or Microsoft with its refocused Cortana product.  It can even make the concept of these voice assistants more relevant to the enterprise market and business customers.

Similarly, telcos and ISPs could create their own voice-driven assistants for use by their customers, typically with functionality that answers what they want out of the telco’s offerings. It can also extend to the hotel and allied sectors that wants to use voice-driven assistants for providing access to functions of benefit to hotel guests like room service, facility booking and knowledge about the local area. Let’s not forget vehicle builders who implement voice-driven assistants as part of their infotainment technology so that the drive has both hands on the wheel and eyes on the road.

This kind of offering can open up a market for the creation of “white-label” voice-assistant platforms that can be “branded” by their customers. As well, some of these assistants can be developed with a focus towards a local market’s needs like high proficiency in a local language and support for local values.

For hardware, the Amazon Voice Interoperability Initiative can open up paths for innovative devices. This can lead towards ideas like automotive applications, smart TVs, build-in use cases like intercom / entryphone or thermostat setups, software-only assistant gateways that work with computers or telephone systems amongst other things.

With the Amazon Voice Interoperability Alliance, there will be increased room for innovation in the voice-driven assistant sector.

Article

You can use the Google Home app to manage how your Google Home smart speaker works during the evening and night hours for a better night’s sleep

Google Home Has A Hidden Night Mode (And We Love It) | Lifehacker

From the horse’s mouth

Google

Manage volume and LED brightness with Night mode (Instruction sheet)

My Comments

Some of you may want to interact with a Google Home smart speaker during the wee hours of the day such as to ask for the time or weather. Or you may touch the device and work it as a night-light. But it can be too bright or loud at these times in a way that people can be woken up at odd hours. Here, some users have to adjust the volume to avoid this risk of disturbance.

But Google has a “Night Mode” feature that allows you to determine the maximum volume and device lighting brightness during times when you don’t want to be disturbed.

Here, you have to use the Google Home mobile-platform app on your smartphone or tablet. As well, your mobile device has be on the same logical network as the Google Home device, which would typically be the same Wi-Fi network when you are thinking of your home network.

When you open the Google Home app, tap on the gear-shape Settings icon and you will see the “Night Mode” setting in the list of settings. There is a toggle switch to enable or disable this mode, and when you enable this mode, the LEDs on the top of your Google Home device will dim while the maximum speaker volume will be softer.

There is an option to schedule the times and days of the week when the Night Mode feature will be active. This may be of importance if you want to make sure it comes in to play on weeknights for example.

There are settings to determine the maximum speaker volume and lighting brightness that will apply to your Google Home smart speaker while it is in the “Night Mode” condition.

The Do Not Disturb option on the Night Mode settings page mutes any notification or system sounds that your Google Home smart speaker makes. But timer or alarm signals will “break through” this setting so you don’t miss that extra alarm you set to wake you up so you can pick up that loved one from the airport as they come off that late flight.

But I am not sure whether these settings can be configured for individual devices or all of the devices bound to the same account. This may be of importance if you want to reduce the volume and lighting brightness on units installed in the bedrooms while one or more units installed in the common living areas are maintained at bright levels; or you want to maintain a common setting across your home.

A feature that can improve on this would be to have the LEDs on top of a Google Home device stay alight but at the maximum brightness to allow you to use the device as a night light. This is more so for those of us who would keep one of these devices in the corridor near the main bathroom or within that bathroom to serve as a “beacon” night-light to enable use of the bathroom at night.

At least Google has provided an option to allow the Google Home device family to work properly without disturbing other people’s sleep at night.