Turns out real people sometimes hear what you say to smart speakers

The revelation that Amazon employs people to listen to voice recordings captured from its Echo devices has apparently surprised some people.

The scoop comes courtesy of Bloomberg and seems to have caught the public imagination, as it has been featured prominently by mainstream publications such as the Guardian and BBC News. Apparently Amazon employs thousands of people globally to help improve the voice recognition and general helpfulness of its smart speakers. That means they have to listen to real exchanges sometimes.

That’s it. Nothing more to see here folks. One extra bit of spice was added by the detail that sometimes workers use internal chatrooms to share funny audio files such as people singing in the shower. On a more serious note some of them reckon they’ve heard crimes being committed but were told it’s not their job to interfere.

Amazon sent Bloomberg a fairly generic response amounting to a justification of the necessity of human involvement in the AI and voice recognition process but stressing that nothing’s more important to it than privacy.

Bloomberg’s main issue seems to be that Amazon doesn’t make it explicit enough that another person may be able to listen into your private stuff through an Echo device. Surely anyone who knowingly installs and turns on a devices that is explicitly designed to listen to your voice at all times must be at least dimly aware that there may be someone else on the other end of the line, but even if they’re not it’s not obvious how explicit Amazon needs to be.

An underlying fact of life in the artificial intelligence era is that the development of AI relies on the input of as much ‘real life’ stuff as possible/ Only be experiencing loads of real interactions and scenarios can a machine learn to mimic them and participate in them. In case there is any remaining doubt, if you introduce a device into your house that is designed to listen at all times, that’s exactly what it will do.

Google wins first round in the battle for the living room

Smart speakers were only about developing a new dynamic in the relationship between the OTTs and the consumer, and Walmart’s new ‘Voice Order’ feature is a taste of things to come.

The new initiative from Walmart is perfect for the Google smart speaker ecosystem, as it plays to the strengths of the internet giant. By simply saying ‘Hey Google, talk to Walmart’ consumers will be able to use their voice to build shopping lists with the grocery mammoth, using any device which has the Google Assistant installed on it.

“With the new voice ordering capabilities we’re building across platforms with partners like Google, we’re helping customers simply say the word to have Walmart help them shop … literally,” said Tom Ward, SVP of Digital Operations at Walmart US.

Of course, the application will not be perfect to start with, but as with anything intelligence related it can be trained and personalised to each individual. At the beginning, users will have to specify what products to put into the cart, but soon enough the virtual assistant will remember these purchases. Saying ‘milk’ won’t put any brand or product into the cart, but the one you bought last time.

This is the futuristic world Silicon Valley had in mind when it started rolling smart speakers out to the world, and we imagine it won’t be too long before the innovation starts catching on.

Although some might suggest Google and Amazon have ambitions to disrupt the audio industry with the launch of their own smart speakers, this was most likely a ploy to drive user acceptance and demonstrate to the mainstream brands there is consumer appetite. If you actually look at the products which Google and Amazon have been championing, they would not compete with the calibre which could be manufactured by the likes of Bose or Bang & Olufsen, but it did start to get consumers using smart speakers.

Google and Amazon are the top-sellers of smart speakers across the world, with Amazon claiming to have now sold more than 100 million products, but the traditional audio giants are starting to release their own products. Sonos is releasing models, so is Samsung. But the traditional audio brands do not have the software smarts to create their own virtual assistants, this is where the likes of Google and Amazon come in.

Sooner or later, smart speakers will be the norm, with the internet giants battling for access to the consumer. A walled garden business model can be created, with the virtual assistant monetizing relationships between the consumer and a third-party. This creates a new dynamic between the consumer and Silicon Valley, offering more opportunities for the internet giants to sell to third-parties, and it looks like Google has won round one in the fight for control of the living room.

Walmart has said other assistants will be available to place orders before too long, but Google was selected as the first partner. This could mean one of two things. Firstly, Google nailed the partnership, commercial elements and technical issues to all for such a feature to be introduced. Then again, it could have paid for the right to be first.

Perhaps it should come as little surprise Google has won the first round here. While Amazon fortunes emerged from hosting an online marketplace and creating a dominant public cloud platform, this sort of feature is true to Google heritage. The Google dominance was created through software, intelligent algorithms and monetizing third-party relationships online. This is nothing more than an extension of this expertise onto a new user interface.

Whichever the case, it is largely irrelevant. Google is now ahead of Amazon when it comes to monetizing the voice user interface. This is a big step forward for the digital economy, and while it might be early days, it does give an indication of the futuristic world we are hurtling towards. With more ‘intelligent’ devices emerging, Google and Amazon could be set to become a lot more powerful and influential.

Google investors slightly spooked by free-spending execs

Revenues might well be booming again at Google, but it seems shareholders are slightly concerned by increased costs, which is one of the fastest growing columns in the spreadsheet.

Looking at the final quarter, revenues stood at $39.3 billion, up 22% year-on-year, though traffic acquisition costs (TAC), what Google pays to make sure it is the dominant search engine across all platforms, operating systems and devices, were up by over $1 billion. Cost-per-click on Google properties were also down. A glimmering ray of sunshine was higher-than-expected seasonal growth for premium YouTube products and services.

Total revenues for 12 months ending December 31 stood at $136.8 billion, up 23% over 2017, while net income was back up to the levels which one would expect at Google, raking in $30.7 billion. The company is not growing as quickly as it used to, while expenses are starting to stack up. Investors clearly aren’t the happiest of bunnies as share price declined 3.1% in overnight trading.

“Operating expenses were $13.2 billion, up 27% year-over-year,” said Alphabet CFO Ruth Porat. “The biggest increase was in R&D expenses, with the larger driver being headcount growth, followed by the accrual of compensation expenses to reflect increases in the valuation of equity in certain Other Bets.

“Growth in Sales and marketing expenses reflect increases in sales and marketing headcount primarily for Cloud and Ads followed by advertising investments mainly in Search and the Assistant.”

Headcount by the end of the last period was up by more than 18,000 employees to 98,771. While CEO Sundar Pichai was keen to point out the business is continuing to invest in improving its core search product, diversification efforts into areas such as the smart speaker market, cloud and artificial intelligence are hitting home. Perhaps investors have forgotten what it’s like to search for the next big idea.

For years, Google plundering the bank accounts with little profit to offer. These days are a long-distant memory, but it is the same for every business which is targeting astronomical growth. You have to perfect the product and then scale. A dip in share price perhaps indicates shareholders have forgotten this concept, but Google is doing the right thing for everyone involved.

Some businesses search for differentiation and diversification when they have to, some do it because they have ambition to remain on top. Those who are searching because they have to are most likely reporting static or declining numbers each month and did not have the vision to see the good days would not last forever. Google is pumping cash into the next idea so when growth in its core business starts to flatten, something else can pick up the slack and pull the business towards more astronomical growth.

This is what is so remarkable about the ‘other bets’ column on the spreadsheets. It might have costs growth every single year, as does the wider R&D column, but having graduated the cloud computing business and most recently Loon, there are businesses which will start to contribute more than they are detracting. This is a company which never sits still, and this is why it is one of the most admired organizations from an entrepreneurial perspective. Shareholders might do well remembering this every now and then.

Looking at joy around the world for the final quarter, US revenues were $18.7 billion, up 21% year-over-year, while EMEA brought in $12.4 billion, up 20% and APAC accounted for $6.1 billion, up 29%. Revenues in LATAM were $2.2 billion, up 16% year-over-year. APAC and LATAM were subject to negative FX fluctuations, particularly in Australia, Brazil and Argentina.

In the specific business units, Google Sites revenues were $27 billion in the quarter, up 22%, with mobile collecting the lion’s share, though YouTube and Desktop contributing growth also. Cloud, Hardware and Play drove the growth in the ‘other’ revenues for Google, collecting $6.5 billion, up 31% year-over-year for the final quarter.

Although these diversification efforts are growing positively, there are also some risks to bear in mind. Firstly, the cloud computing business is losing pace with Microsoft and AWS. Google is making investments to attempt to buy its way through the chasm, but it will be tough going as both these businesses make positive steps forward also.

Secondly, some properties and developers are choosing to circumnavigate the Google Play Store, instead taking their titles direct to the consumer. This is only a minor segment of the pie for the moment and there will be a very small proportion of the total who actually have the footprint to do this (Fortnite for example), though it is a trend the team will want to keep an eye on. Perhaps the 30% commission Google charges developers will be reconsidered to stem dissenting ideas.

Finally, the data sharing economy which will sit behind the smart speaker and smart home ecosystem is facing a possible threat. Google will not make the desired billions from hardware sales, but it will from the operating systems and virtual assistant powering the devices. Collecting referral fees and connecting buyers with sellers is what Google does very well, though this business model might be under threat from new data protection and privacy regulations.

The final one is not just a challenge to the potential billions hidden between the cushions in the smart home’s virtual sofa, but the entire internet economy. GDPR complaints are currently being considered and potential consequences to how personal data is collected, processed and stored are already being considered. The Google lawyers will have to be on tip-top form to minimise the disruption to the business, and wider data sharing economy.

Costs might be up and while there are dark clouds on the horizon, Pichai and his executives are moving in the right direction. The lawyers can lesson the potential impact of regulation, but the exploration encouraged by the management team in the ‘other bets’ segment is what will fuel Google in the future. Costs should be controlled, but spending should also be encouraged.

Deloitte predicts 50k 5G smartphone in the UK by 2019-end

While the vast majority will have to wait some time before experiencing the euphoria of an extra ‘G’ Deloitte is predicting there will 50,000 early adopters in the UK.

After several years of slugging, the glorious 5G world is upon us. First in the US and South Korea, though pockets are starting to emerge everywhere else as well. San Marino is live while it won’t be long before countries like China and Japan start hitting the green button.

“The introduction of 5G handsets expected this year will look a lot like 2010, when 4G phones first entered the market,” said Dan Adams, Head of Telecommunications at Deloitte.

“There will be a lot of noise in the first year from vendors vying to be first to market, and relatively little action from consumers. We’re not talking about an overnight switch to faster connectivity with lower latency, we will see 5G used by consumers in hotspot locations in the next two to three years, with mass adoption by 2025.”

The first devices are likely to be with us in Q2, though this year’s Mobile World Congress will almost certainly be a shouting contest between the main smartphone manufacturers. It’s already rumoured Samsung will be launching a foldable-phone (albeit not 5G) prior to the event, while LG and Motorola are also in the running to produce a 5G compatible phone.

In total, Deloitte predicts roughly 20 handset brands will launch 5G-ready handsets across 2019, with shipments totalling one million. This is still a tiny fraction of the 1.5 billion smartphones which will be sold through the year, though 50,000 of them could be heading to the UK.

Looking at the networks, there might not be much to choose from across the UK. EE has confirmed it will launch 5G across 16 cities in 2019, though these will only be in the busiest locations. Vodafone will also launch this year, though it is being coy as to when. Three is telling the same story, while O2 has confirmed its customers will have to wait until 2020. One thing is clear, these will be incredibly limited deployments and it will be years until coverage reaches what the demanding user would consider adequate.

Whether this justifies the hype, or the extortionate amount handset manufacturers will inevitably charge the glory-seekers for the new devices, we’ll leave you to decide, but it will take years for the devices to be considered mainstream. Deloitte expects worldwide 5G smartphone sales to represent 1% of the total smartphone sales by the end of 2020, with 2-3 million Brits getting their hands on the devices. As Adams points out above, 2025 is when the team expect 5G devices to hit mass adoption.

Another interesting growth area the Deloitte team is keeping an eye on is the smart speakers segment.

“Smart speaker adoption has seen phenomenal growth in recent years,” said Paul Lee, Global Head of Research for TMT at Deloitte

“With improvements continuing to be made, demand for smart speakers could be in the many billions of units, possibly even higher than for smartphones. In the future, smart speakers have the potential to be installed in every room in a house, hotel, office, school and even beside every hospital bed.”

Smart speakers are the flashy product which will attract a lot of the consumer market, but the power of the virtual assistants is what could take the segment to the next level. We’ve long anticipated the breakthrough of artificial intelligence in the workplace, but perhaps the slightly sluggish resistance has been down to the delivery model of the applications.

Should smart speakers be adopted in hotel rooms, hospitals and offices in the way which Deloitte anticipates, the world is opened up for industry specific applications of virtual assistants. One area which might help this adoption is the price point.

While smart speakers were initially an expensive appliance for the home, the normalisation of the product in the eyes of the consumer has peaked the interest of traditional consumer electronics manufacturers. With more manufacturers, including those with the ability to produce goods at greater scale, entering the fray competition will increase, bringing prices down, while advertising will also grow, fuelling interest in the bellies of the consumer.

Deloitte anticipates the marker for internet-connected speakers with integrated digital assistants will be increase to £5.6 billion in 2019, selling 164 million units at an average selling price of £34. This would represent a 63% growth rate, making smart speakers the fastest-growing connected device category worldwide, leading to an installation base of more than 250 million units by the end of the year.

This is a price point which would make enterprise adoption of the devices more interesting, and as time moves on, it will get cheaper. The increased introduction of industry-specific virtual assistant and AI applications will certainly help this segment also.

After years of promises and false-dawns, 2019 might prove to be a blockbuster year after all. There’s still a lot which could go wrong, but here’s to hoping.

Orange squares up to OTTs in battle for smart home

With it abundantly clear connectivity alone is not enough to meet the profitability ambitions of the telcos, Orange has made a fresh push to wrestle control of the smart home away from those greedy internet players.

As expected, the Orange and Deutsche Telekom partnership has flourished into a diversification venture. For Orange, the team will be launching the Djingo Speaker, while DT will be launching its own version branded the Magenta Speaker. These are fundamentally the same product, operated by the same AI, tailored for the individual markets.

New products should hardly come as a surprise, but this is the tip of the spear aimed directly at the useful ecosystem created by Amazon and Google in the smart home.

Originally the router might have appeared to be the logical focal point of the fabled smart home ecosystem, though inactivity from the telcos and aggressive deployment from the OTTs has seen this shift to the smart speaker. Orange’s Live Box will seemingly be the home of the connected services portfolio, though with the smart speaker the team now have an interface which is increasingly becoming normalised with the consumer.

The big question is whether Orange is able to demonstrate the value of itself as a provide to the customer, above and beyond what the OTTs can offer. And they are playing an interesting angle.

“The business model is based on the subscription charge,” said Stephane Ricard, Orange CEO. “We won’t squeeze your data to make money.”

With the world quickly turning against the data-sharing economy, thanks to governments gradually exposing the complicated nature of the data machine, this might be a useful statement to make. Orange will charge a subscription for the added value services, not use personal information as a commodity in the manner the OTTs are.

This is where Orange might find its first challenge. Less than one in four French citizens have a connected device today, aside from a smartphone, while only 10% make use of a digital assistant. The initial connected life services offered by Orange, linking up everything from connectivity to banking and entertainment, will be free, though monetization depends on luring customers to the more premium services, such as security.

This is one of the services which will be placed on top of the connected ecosystem created by Orange in the smart home. Working in partnership with Groupama, the Protected Home is a security solution which can be remotely managed by the user. There will be future joint ventures and solutions launched on top of the smart home ecosystem, but Orange needs to convince the user it is worth it in the first instance.

With low penetration of connected devices and virtual assistants, the French clearly aren’t that enthused by consumer IOT right now. Perhaps the Orange brand, a trusted and credible company in France, can change this image. Though whether it can compete toe-to-toe with the likes of Google and Amazon in developer power, remains to be seen. We doubt it, though perhaps the world does not need an overly complex virtual assistant right now.

Users don’t need a virtual assistant to restock fridges, or arrange meetings, they just a link to control the smart home. This is an area which the user is still getting used to, therefore perhaps its inability to create an overly-complex and super-intelligent virtual assistant will work in its favour?

Orange’s ambition is to be a multi-service vendor with connectivity acting as the bridge between various different services. The banking venture is now a year old, making steady progress, and the team will hope the same success can be replicated in the smart home segment.

Google adds some Pixels

Internet giant Google ramped up its involvement in the consumer hardware space with the launch of new Pixel branded smartphones and tablets as well as a home hub.

The Pixel 3 and its XL variant offer both an industrial design and spec upgrade on their predecessors. Initial impressions indicate the redesign is well received and the spec upgrades are significant. There also seems to be more AI stuff going on, including a call screening functions that taps into Duplex technology to save you having to interact with a caller if you’re not sure about them.

Google debuted a new device category in the form of the Pixel Slate – a tablet running Chrome OS that seems to be positioned as a direct competitor to Microsoft’s Surface product range, with an emphasis on hybrid laptop functionality. Once more initial takes seem positive, especially about its attempt to be the best of both worlds, although the full range of requisite peripherals and accessories does make it an expensive proposition.

Lastly we have the Home Hub, which is an AI-driven smart speaker with a 7-inch screen that will compete with equivalent products from Amazon and Facebook. One big difference is that Google is making a virtue of it not having a camera installed in an apparent bid for people to take it into the bedroom or even the bog. There’s also a physical mute switch to prevent the device listening to you, which seems like a good say to allay fears about being spied on by Google, but does call into question what the point of the device is.

“Our goal with these new products, as always, is to create something that serves a purpose in people’s lives – products that are so useful they make people wonder how they ever lived without them,” said Rick Osterloh, VP of Hardware at Google. “The simple yet beautiful design of these new devices continue to bring the smarts of the technology to the forefront, while providing users with a bold piece of hardware.”

The Pixel 3 starts at £739, with the XL coming in at £869. The Slate starts at £549 without peripherals, while the Home Hub will set you back £139. Google has managed to throw down the gauntlet to the majority of the consumer tech world with one set of launches, which is fun, but time will tell whether any of them are able to claim significant market share. Here’s a vid.

 

Smart speakers starting to drive next era of digital

Years usually pass before the world realises a technology breakthrough actually happened and it catches on; the voice user interface might just have arrived at that watershed moment.

The fantastic breakthrough of touchscreen mobile phones is often attributed to Apple, though if you go back to 1992 IBM unveiled a phone called Simon which featured the first touchscreen. Apple didn’t invent the concept of touch screens, it simply innovated, making the iPhone a genuine smartphone as opposed to a PDA. This might seem like an odd introduction, but the same trend is emerging in the smart speaker world.

Amazon and Google did not invent the concept of the voice user interface, they simply used their brands to effect change and offer a product which was dutifully adopted by the masses which call themselves fans. In releasing their own smart speakers, the two internet giants did what other companies couldn’t; they normalised the voice user interface.

According to Nielsen’s MediaTech Trender survey, the smart speaker has penetrated the mass market and is normalising the concept of the smart home, as well as the idea of your voice being the control function. Across the US, 40% of homes now own at least one smart home device, with 24% owning a smart speaker, up from 22% in the previous quarter. Of those who currently own a smart home device, 65% plan to purchase more. Looking at the speakers themselves, usage is up, the average user interacts with the device for 72 minutes on the weekend and 65 minutes during the week, while 81% of users report using voice-command searches for real-time information, such as weather and traffic conditions, during a typical week. The more normal it becomes to use your voice in the home, the more acceptable it becomes elsewhere in the world.

Another interesting statistic from the report are the services synced to the speakers. Music streaming services are unsurprisingly the most popular, 53%, while the second most popular is shopping apps at 52%. With the user seemingly becoming accustomed to ordering goods through the smart speaker, there are a horde of new opportunities emerging, from grocery shopping to on-demand purchases linked to advertisements.

Finally, the most device synced to the smart speakers is the smartphone. This might seems like a very obvious statement, though only 32% of the respondents have linked their smartphone to the device. This is a small percentage of what is possible, though the potential to learn more about these individuals who have synced their devices is quite exciting. The virtual assistant is no-longer limited to the users home and can start to learn about habits in the big, wide world. This offers a much more in-depth opportunity to create valuable, personalised services.

As it stands, the smart speaker is little more than an entertainment product. 90% of users listen to music on the devices, 81% search for real-time information such as the weather, 68% listen to the news and 68% use it for alarms or timers. However, these devices are introducing new concepts and features which are gradually becoming accepted and normalised by the user. The voice user interface is an incredibly important one.

Just like the touch interface opened up new opportunities to make money, the voice interface will do the same. But this is a while down the road, mass adoption of both new devices and the normalisation of new concepts need to take place first. New ideas open the mind up to even more new ideas, including services and products, as well as blurring the lines of what would be considered intrusive or unacceptable. The smart speaker is playing a critical role here.

Alexa is starting to turn into a genuinely helpful assistant

Amazon has unveiled a host of new features and skills for its Alexa virtual assistant, edging the living room closer to the intelligent dream we’ve all been promised.

While you cannot argue with the gimmicky entertainment brought by virtual assistants, you have to wonder whether it is anything more than five minutes of entertainment or making our lives easier in the very smallest (and often irrelevant) of ways. The new features and skills released by Amazon are starting to add some clarity to the smart home as we all imagine it from watching too many re-runs of Back to the Future 2.

“The Alexa service is always getting smarter, whether you’re using the Echo you bought three years ago or an Echo Show you buy tomorrow,” said said Tom Taylor, Senior VP of Amazon Alexa. “We have thousands of engineers and scientists inventing on behalf of customers, and today we’re excited to introduce even more features to help make customers’ lives simpler, safer, and more convenient.

“Soon customers will be able to manage their email, easily secure their home, watch the shows they love on Echo Show, and make their daily routines more productive – all just by asking Alexa.”

Right now virtual assistants are very limited in the way they work. This is partly due to customers not utilising the capabilities to full potential, though the breadth of features and skills does need to be fleshed out. It might be cruel to point the finger at Amazon, it is still early days after all, but with the big promises made in advertisements, the virtual assistants are a bit drab. That said, some of the new features do look pretty good. The difference is underlying interaction with other applications and features.

Take the new location based reminders. It’s a simple idea, but linking reminders up with GPS adds value. How many times have you walked home from the tube station, only to realise you forgot to buy peas when you are half-way through cooking your dinner. Now you can ask Alexa to remind you to pick up peas, post a letter, drop off the dry cleaning or buying a last minute birthday present, when you’re passing by the relevant establishment.

The routines is another area which been improved as well. This is an interesting feature which can be adapted to each individual. The morning playlist might depend on the day for instance, or lights are triggered depending on motion and your routine. Both of these examples take the virtual assistant away from the simple command-action scenario and factor in other variables which are not dependent on proactive actions from the user. It is actually starting to become smart.

Later in the year you’ll start to see some even more interesting features with Alexa actually making sensible suggestions depending on your actions and commands. For example, if you activate the bedtime routine by saying ‘Good night Alexa’, the white noise playlist will kick in, and Alexa might ask you whether you want it to switch off the living room light you left on. Features like this will make the virtual assistant much more than a gimmick.

The next step will be deeper integration with other applications such as Outlook calendars. When Alexa prompts you to change your alarm the night before because it has spotted an early morning meeting, it’ll start to be a genuine assistant. One step further would be linking to weather and travel update services so it can proactively change the alarm in the middle of the night if it decides your commute will take longer than it usually should.

The promise of virtual assistants has been very glorious, and so far it hasn’t met the expectations. But updates like this are making Alexa an genuinely helpful and interesting proposition.