There aren’t any good stock photos of “ten thousand”, so this piece will just have lots of kittens
I
started writing a blog in May 2016, partly because I kept writing rants
on Facebook that apparently were “too good not to be online somewhere”,
and partly because I was bored after my Master’s degree and wanted
something to do with my Sunday mornings.
Sleeping in, of course, was never an option.
This is Luna. Luna is my 6am alarm clock. Every. Single. Day
18
months later, and I’ve written about 100,000 words, been published in
all sorts of places, and am now getting regular offers to pitch to major
publications — more on this in the coming months.
And
most importantly of all, I got to 10,000 followers. This time last
year, it was 100 and about half of them were related to me.
All in all, it’s been a good year.
Pictured: Getting what you always wanted
So
what’s in store for the Health Nerd? You’ll be happy to know that this
year I’ve applied for a PhD with the University of Wollongong, which is
actually super exciting and not scary like it feels to me sometimes. I’m
also going to be — hopefully — releasing some episodes of a podcast
that I’ve started with a brilliant co-host. The topic will be science in
the media and I’m really excited to introduce all of you to my dulcet
tones over the airwaves.
I’m so much less awkward than I am in text.
What
does all of this activity mean to the blog? Nothing! I’ll still be
aiming for my regular one health story a week on Medium, as well as an
extra member’s-only article a month for all you subscribers who love
that extra content.
Pictured: “Extra content”
To
sum up, I’d just like to say thank you to you all. I’d never have made
it here without all you brilliant people following me and making this
all worthwhile. It was a fantastic 2017, and 2018 shows every sign of
being brilliant as well.
We wanted to put the Museum of English Rural Life
(The MERL) on Google Streetview to make us more accessible to those
with Autism Spectrum Disorder (ASD). For people with ASD it helps to
know what to expect at a place before they arrive, and Google Streetview
remains one of the most popular ways of scoping a place out. (Our offer
for people with ASD is forthcoming at the MERL.)
We thought it would be difficult to get on Streetview. What we didn’t realise is that:
pretty much anyone with the right equipment can put themselves on Google Streetview
it isn’t rocket science
So,
in this blog I’m going to tell you how we did it, in case you also want
to do it. If you want to skip straight to our Google Streetview tour, click here.
The background
In case you don’t know, Google Streetview
is attempting to capture every street in 360-degree photography. You
just drag the little yellow guy on Google Maps onto the street and have a
look.
The closest you could previously get in Streetview (hey, that’s my bike!)
Google Streetview also extends inside buildings, for which you used to have to hire a Trusted Pro to photograph your building. Google now allows anyone to do it themselves, kind of like what crowd-sourcing Panoramio used to be except for 360° photos.
The equipment
Google
will accept any photos taken with decent 360° cameras, and even accepts
photo spheres made with a normal smartphone camera if they’re good
enough. They have a very good page on how to publish for Google Streetview here.
These people are in exciting places but most of the 360 photos I’ve seen are of ducks on the canal and nail parlours
So,
technically you just need a smartphone, but the photo-spheres I’ve made
using just a normal camera almost always come out a bit glitchy. So I
suggest getting a real camera.
The
Theta-S takes images using two cameras on either side of its body, then
stitches them together for you. You just export the jpeg and upload it
onto something like Google Streetview or Facebook – sites which can
translate the file into an interactive photo-sphere.
Setting up the tour
So taking 360° photos is literally as easy as pressing a button, but planning the actual 360° tour? Not so much.
For
starters we didn’t want the photographer in the photo, so we mounted
the Theta S on a monopod and hid behind walls as we took the photo. This only failed once.
We
decided to capture the Museum while it was empty and shot on a Monday,
our closure day. Images of an empty museum, however, may give the wrong
impression of the museum to someone with ASD, as we usually have
visitors milling around. We plan to test this out with focus groups.
The ground floor had more than enough photos.
We
also wanted to be able to capture the whole museum, and planning our
tour was made easier by how our galleries are fairly one-way and linear.
Because
we only have our ground floor layer on Google Maps, though, we had to
miss out our first floor open store. We originally had both ground and
first floors published, but rapidly realised it was confusing people as
they kept switching randomly between floors in Streetview. We hope to
see whether getting our ground and first floor plans published on Google means we can then separate Streetview tours between them.
GO INSIDE WITH INDOOR MAPS WHENEVER GOOGLE CAN BE ARSED TO ACTUALLY UPLOAD YOUR SUBMISSION
Taking photos
Google
suggests taking photos a metre apart indoors, but we rarely kept to
this. On our first run we had a distance of something like five metres,
and then we went back to fill in some gaps.
There’s
an option to connect the Google Streetview app to your 360° camera, but
we chose to take the photos and upload them separately (Import 360°
photos). I highly suggest taking all the photos you need, cut any
mistakes and then upload them all in one batch. If you have a museum the
size of the MERL you can do the whole museum in one go (76 photos), or
if you’re larger you could do it by gallery.
After
taking our photos we also realised some of them featured copyrighted
artworks. We opened these images in Photoshop, blurred out the artworks
and re-saved them – they still worked fine after editing, which was a
relief. The Google Streetview app also gives you the option of
automatically blurring faces.
Publishing
Once you have your photos collected you need to select all your photos and attach them to an address (i.e., your museum).
With
all the photos still selected, you then need to choose their precise
locations on Google Maps. This step is probably the most time-consuming.
As well as placing them in the exact spot you took them on your
floorplan, you also need to orient them to the compass so they’re
pointed in the right direction. This is very important for when you
connect your photos in a tour.
When
your photos are placed and oriented you can publish them to Google
Streetview. They usually show up fairly fast on the app and on desktop.
Connecting photos
The
beauty of Streetview is that you can place your photos in a sequential
tour. The option to link photos is only available after publication.
To do this I’d again suggest selecting all your photos at once, and then choosing the option to place and link.
You connect your photos by simply tapping the line between them, and you can link more than one picture to another.
That’s it.
It
updates instantly on the app, but it takes a couple of days before you
will be able to navigate through your photos on desktop using your
keyboard’s arrow keys or on your phone by tapping around.
The
aim of publishing our museum on Google Streetview is to prepare people
for what to expect at the Museum. It definitely accomplishes that.
We
considered photos and video, and have these options available too, but
nothing beats Streetview for giving the full picture. People already use
Google and Streetview, and it meant we could also embed the tour on our
website.
With
our planning, testing and re-runs the whole process probably took us
three full days of work. If you know what you need to capture, organise a
day for photography and dedicate the day to editing the photos then you
could easily get a museum the size of the MERL done in a day’s work.
A note on the Google Streetview app
I
don’t know whether it’s because I installed it on an iPad, but the
Google Streetview app is buggy as hell. It crashes, it is unresponsive
and often the map is completely obscured by cards. Prepare to be
frustrated, and work/save in batches to avoid losing your work.
Another
weird glitch which hasn’t been fixed yet is the option to transfer the
rights of your photos to the place where you took them. This is
primarily intended for Trusted Pros who are hired to make 360-degree
tours, and who then transfer the rights to the people who commissioned
the tour. It seemed strange that we could transfer rights to photos
taken using the MERL Google account to our same Google account tied to
the business. We did it anyway and all of our photos promptly
disappeared from Google Maps.
So, don’t do that until they’ve fixed it? But otherwise have fun.
A
prankster who made repeated hoax distress calls to the US Coast Guard
over the course of 2014 probably thought they were untouchable. They
left no fingerprints or DNA evidence behind, and made sure their calls
were too brief to allow investigators to triangulate their location.
Unfortunately
for this hoaxer, however, voice analysis powered by AI is now so
advanced that it can reveal far more about you than a mere fingerprint.
By using powerful technology to analyse recorded speech, scientists
today can make confident predictions about everything from the speaker’s
physical characteristics — their height, weight, facial structure and
age, for example — to their socioeconomic background, level of income
and even the state of their physical and mental health.
One of the leading scientists in this field is Rita Singh of Carnegie Mellon University’s Language Technologies Institute.
When the US Coast Guard sent her recordings of the 2014 hoax calls,
Singh had already been working in voice recognition for 20 years. “They
said, ‘Tell us what you can’,” she told the Women in Tech Show podcast earlier this year. “That’s when I started looking beyond the signal. How much could I tell the Coast Guard about this person?”
Rita Singh is an expert in speech recognition
What your voice says about you
The
techniques developed by Singh and her colleagues at Carnegie Mellon
analyse and compare tiny differences, imperceptible to the human ear, in
how individuals articulate speech. They then break recorded speech down
into tiny snippets of audio, milliseconds in duration, and use AI
techniques to comb through these snippets looking for unique
identifiers.
Your
voice can give away plenty of environmental information, too. For
example, the technology can guess the size of the room in which someone
is speaking, whether it has windows and even what its walls are made of.
Even more impressively, perhaps, the AI can detect signatures left in
the recording by fluctuations in the local electrical grid, and can then
match these to specific databases to give a very good idea of the
caller’s physical location and the exact time of day they picked up the
phone.
This
all applies to a lot more than hoax calls, of course. Federal criminal
cases from harassment to child abuse have been helped by this relatively
recent technology. “Perpetrators in voice-based cases have been found,
have confessed, and their confessions have largely corroborated our
analyses,” says Singh.
Portraits in 3D
And
they’re just getting started: Singh and her fellow researchers are
developing new technologies that can provide the police with a 3D visual
portrait of a suspect, based only on a voice recording. “Audio can us
give a facial sketch of a speaker, as well as their height, weight,
race, age and level of intoxication,” she says.
But
there’s some way to go before voice-based profiling technology of this
kind becomes viable in a court. Singh explains: “In terms of
admissibility, there will be questions. We’re kind of where we were with
DNA in 1987, when the first DNA-based conviction took place in the
United States.”
This
has all proved to be bad news for the Coast Guard’s unsuspecting
hoaxer. Making prank calls to emergency services in the US is regarded
as a federal crime, punishable by hefty fines and several years of jail
time; and usually the calls themselves are the only evidence available.
Singh was able to produce a profile that helped the Coast Guard to
eliminate false leads and identify a suspect, who they hope to bring a
prosecution soon.
Given
the current exponential rate of technological advancement, it’s safe to
say this technology will become much more widely used by law
enforcement in the future. And for any potential hoax callers reading
this: it’s probably best to stick to the old cut-out newsprint and glue
method for now. Just don’t leave any fingerprints.
Voice is the future. The world’s technology giants are clamoring for vital market share, with ComScore projecting that “50% of all searches will be voice searches by 2020.”
However,
the historical antecedents that have led us to this point are as
essential as they are surprising. Within this report, we take a trip
through the history of speech recognition technology, before providing a
comprehensive overview of the current landscape and the tips that all
marketers need to bear in mind to prepare for the future.
The History of Speech Recognition Technology
Speech
recognition technology entered the public consciousness rather
recently, with the glossy launch events from the tech giants making
worldwide headlines.
The appeal is instinctive; we are fascinated by machines that can understand us.
From
an anthropological standpoint, we developed the spoken word long in
advance of its written counterpart and we can speak 150 words per
minute, compared with the paltry 40 words the average person can type in
60 seconds.
In
fact, communicating with technological devices via voice has become so
popular and natural that we may be justified in wondering why the
world’s richest companies are only bringing these services to us now.
The
history of the technology reveals that speech recognition is far from a
new preoccupation, even if the pace of development has not always
matched the level of interest in the topic. As we can see below, major
breakthroughs dating back to the 18th century have provided the platform
for the digital assistants we all know today.
Created by author
The
earliest advances in speech recognition focused mainly on the creation
of vowel sounds, as the basis of a system that might also learn to
interpret phonemes (the building blocks of speech) from nearby
interlocutors.
These
inventors were hampered by the technological context in which they
lived, with only basic means at their disposal to invent a talking
machine. Nonetheless, they provide important background to more recent
innovations.
Dictation
machines, pioneered by Thomas Edison in the late 19th century, were
capable of recording speech and grew in popularity among doctors and
secretaries with a lot of notes to take on a daily basis.
However,
it was not until the 1950s that this line of inquiry would lead to
genuine speech recognition. Up to this point, we see attempts at speech
creation and recording, but not yet interpretation.
Audrey,
a machine created by Bell Labs, could understand the digits 0–9, with a
90% accuracy rate. Interestingly, this accuracy level was only recorded
when its inventor spoke; it hovered between 70% and 80% when other
people spoke to Audrey.
This
hints at some of the persistent challenges of speech recognition; each
individual has a different voice and spoken language can be very
inconsistent. Unlike text, which has a much greater level of
standardization, the spoken word varies greatly based on regional
dialects, speed, emphasis, even social class and gender. Therefore,
scaling any speech recognition system has always been a significant
obstacle.
Alexander
Waibel, who worked on Harpy, a machine developed at Carnegie Mellon
University that could understand over 1,000 words, built on this point:
“So
you have things like ‘euthanasia’, which could be ‘youth in Asia’. Or
if you say ‘Give me a new display’ it could be understood as ‘give me a
nudist play’.”
Until
the 1990s, even the most successful systems were based on template
matching, where sound waves would be translated into a set of numbers
and stored. These would then be triggered when an identical sound was
spoken into the machine. Of course, this meant that one would have to
speak very clearly, slowly, and in an environment with no background
noise to have a good chance of the sounds being recognized.
IBM
Tangora, released in the mid-1980s and named after Albert Tangora, then
the world’s fastest typist, could adjust to the speaker’s voice. It
still required slow, clear speech and no background noise, but its use
of hidden Markov models allowed for increased flexibility through data
clustering and the prediction of upcoming phonemes based on recent
patterns.
Although
it required 20 minutes of training data (in the form of recorded
speech) from each user, Tangora could recognize up to 20,000 English
words and some full sentences.
The
seeds are sown here for voice recognition, one of the most significant
and essential developments in this field. It was a long-established
truism that speech recognition could only succeed by adapting to each
person’s unique way of communicating, but arriving at this breakthrough
has been much easier said than done.
It
was only in 1997 that the world’s first “continuous speech recognizer”
(ie. one no longer had to pause between each word) was released, in the
form of Dragon’s NaturallySpeaking software. Capable of understanding
100 words per minute, it is still in use today (albeit in an upgraded
form) and is favored by doctors for notation purposes.
Machine
learning, as in so many fields of scientific discovery, has provided
the majority of speech recognition breakthroughs in this century. Google
combined the latest technology with the power of cloud-based computing
to share data and improve the accuracy of machine learning algorithms.
This culminated in the launch of the Google Voice Search app for iPhone in 2008.
Driven
by huge volumes of training data, the Voice Search app showed
remarkable improvements on the accuracy levels of previous speech
recognition technologies. Google built on this to introduce elements of
personalization into its voice search results, and used this data to
develop its Hummingbird algorithm, arriving at a much more nuanced
understanding of language in use. These strands have been tied together
in the Google Assistant, which is now resident on almost 50% of all
smartphones.
It
was Siri, Apple’s entry into the voice recognition market, that first
captured the public’s imagination, however. As the result of decades of
research, this AI-powered digital assistant brought a touch of humanity
to the sterile world of speech recognition.
After
Siri, Microsoft launched Cortana, Amazon launched Alexa, and the wheels
were set in motion for the current battle for supremacy among the tech
giants’ respective speech recognition platforms.
In
essence, we have spent hundreds of years teaching machines to complete a
journey that takes the average person just a few years. Starting with
the phoneme and building up to individual words, then to phrases and
finally sentences, machines are now able to understand speech with a
close to 100% accuracy rate.
The
techniques used to make these leaps forward have grown in
sophistication, to the extent that they are now loosely based on the
workings of the human brain. Cloud-based computers have entered millions
of homes and can be controlled by voice, even offering conversational
responses to a wide range of queries.
That journey is still incomplete, but we have travelled quite some distance from the room-sized computers of the 1950s.
The Current Speech Recognition Landscape
Smartphones
were originally the sole place of residence for digital assistants like
Siri and Cortana, but the concept has been decentralized over the past
few years.
At
present, the focus is primarily on voice-activated home speakers, but
this is essentially a Trojan horse strategy. By taking pride of place in
a consumer’s home, these speakers are the gateway to the proliferation
of smart devices that can be categorized under the broad ‘Internet of
Things’ umbrella. A Google Home or Amazon Echo can already be used to
control a vast array of Internet-enabled devices, with plenty more due
to join the list by 2020. These will include smart fridges, headphones,
mirrors, and smoke alarms, along with an increased list of third-party
integrations.
Recent Google research
found that over 50% of users keep their voice-activated speaker in
their living room, with a sizeable number also reporting that they have
one in their bedroom or kitchen.
And
this is exactly the point; Google (and its competitors) want us to buy
more than one of these home devices. The more prominent they are, the
more people will continue to use them.
Their
ambition is helped greatly by the fact that the technology is now
genuinely useful in the accomplishment of daily tasks. Ask Alexa, Siri,
Cortana, or Google what the weather will be like tomorrow and it will
provide a handy, spoken summary. It is still imperfect, but speech
recognition has reached an acceptable level of accuracy for most people
now, with all major platforms reporting an error rate of under 5%.
As
a result, these companies are at pains to plant their flag in our homes
as early as possible. Hardware, for example in the shape of a home
speaker system, is not something most of us purchase often. For example,
if a consumer buys a Google Home, it seems probable that they will
complement this with further Google-enabled devices, rather than
purchase from a rival company and create a disjointed digital ecosystem
under their roof. Much easier to seek out devices that will enable
continuity and greater convenience.
For this reason, it makes sense for Amazon to sell the Echo Dot for as low as $29.99. That equates to a short-term financial loss for Amazon on each device sold, but the long-term gains will more than make up for it.
There are estimated to be 33 million smart speakers in circulation already (Voice Labs report, 2017) and both younger and older generations are adopting the technology at a rapid rate.
In
fact, the demographics of an assistant “superuser,” someone who spends
twice the amount of time with personal assistants on a monthly basis
than average — is a 52-year old woman, spending 1.5 hours per month with
assistant apps.
Perhaps
most importantly for the major tech companies, consumers are
increasingly comfortable making purchases through their voice-enabled
devices.
Google
reports that 62% of users plan to make a purchase through their speaker
over the coming month, while 58% use theirs to create a weekly shopping
list:
Short-term
conclusions about the respective business strategies of Amazon and
Google, in particular, are relatively easy to draw. The first-mover
advantage looks set to be marked in this arena, especially as speech
recognition continues to develop into conversational interactions that
lead to purchases.
We have written before
about the two focal points of the voice search strategy for the tech
giants: the technology should be ubiquitous and it must be seamless.
Voice is already a multi-platform ecosystem, but we are some distance
from the ubiquity it seeks.
To
gain insight into the likely outcome of the current competition, it is
worth assessing the strengths and weaknesses of the four key players in
western markets: Amazon, Google, Apple, and Microsoft.
Amazon
First-party Hardware: Echo, Echo Dot, Echo Show, Fire TV Stick, Kindle.
Digital Assistant: Alexa
Usage Statistics:
“Tens of millions of Alexa-enabled devices” sold worldwide over the 2017 holiday season (Amazon)
75% of all smart speakers sold to date are Amazon devices (Tech Republic)
The
Echo Dot was the number one selling device on Amazon over the holidays,
with the Alexa-enabled Fire TV stick in second place. (Amazon)
The
average Alexa user spends 18 minutes a month interacting with the
device, compared to just five minutes for Google Home (Gartner)
There are now over 25,000 skills available for Alexa (Amazon)
Overview:
The
cylindrical Echo device and its younger sibling, the Echo Dot, have
been the runaway hit of the smart speaker boom. By tethering the
speakers to a range of popular third-party services and ‘skills’, Amazon
has succeeded in making the Echo a useful addition to millions of
households.
As Dave Limp, head of Amazon devices, put it recently,
“We think of it as ambient computing, which is computer access that’s less dedicated personally to you but more ubiquitous.”
Ubiquity seems a genuine possibility, based on the sales figures.
After
a holiday season when the Echo Dot became the most popular product on
Amazon worldwide, the Alexa app occupied top position in the App Store,
ahead of Google’s rival product.
Amazon’s
heritage as an online retailer gives it an innate advantage when it
comes to monetizing the technology, too. The Whole Foods acquisition
adds further weight to this, with the potential to integrate the offline
and online worlds in a manner other companies will surely envy.
Moreover,
Amazon has never depended on advertising to keep its stock prices
soaring. Quite the contrary, in fact. As such, there is less short-term
pressure to force this aspect of their smart speakers.
With
advertisers keen to find a genuine online alternative to Google and
Facebook, Amazon is in a great position to capitalize. There is a fine
balancing act to maintain here, nonetheless. Amazon has most to lose, in
terms of consumer trust and reputation, so it will only move into
advertising for Alexa carefully.
The company denies it has plans to do so, but as research company L2 Inc wrote recently,
Amazon
has approached major brands asking if they would be willing to pay for
Amazon’s Choice, a designation given to best-in-class products in a
particular category.
We
should expect to see more attempts from Amazon to provide something
beyond just paid ads on search results. Voice requires new advertising
solutions and Amazon will tread lightly at first to ensure it does not
disrupt the Alexa experience. The recently announced partnership with publishing giant Hearst is a sign of things to come.
The
keys to Alexa’s success will be the integration of Amazon’s own assets,
along with the third-party support that has already led to the creation
of over 25,000 skills. With support announced
for new headphones, watches, fridges, and more, Amazon looks set to
stay at the forefront of voice recognition technology for some time to
come.
Google
First-party Hardware: Google Home, Google Home Mini, Google Home Max, Pixelbook, Pixel smartphones, Pixel Buds, Chromecast, Nest smart home products.
Digital Assistant: Google Assistant
Usage Statistics:
Google Home has a 24% share of the US smart speaker market (eMarketer)
There are now over 1,000 Actions for Google Home (Google)
Google Assistant is available on over 225 home control brands and more than 1,500 devices (Google)
The most popular Google Assistant apps are games, followed closely by home control applications (Voicebot.ai)
Overview:
Google
Assistant is directly tied to the world’s biggest search engine,
providing users with direct access to the largest database of
information ever known to mankind. That’s not a bad repository for a
digital assistant to work with, especially as Google continues to make
incremental improvements to its speech recognition software.
Recent
research from Stone Temple Consulting across 5,000 sample queries found
Google to be the most accurate solution, by quite some distance:
Combined
with Google Photos, Google Maps, YouTube, and a range of other
effective services, Google Assistant has no shortage of integration
possibilities.
Google
may not have planned to enter the hardware market again after the
lukewarm reception for its products in the past. However, this new
landscape has urged the search giant into action in a very serious way.
There is no room for error at the moment, so Google has taken matters
into its own hands with the Pixel smartphones, the Chromecast, and of
course the Home devices.
The
Home Mini has been very popular, and Google has added the Home Max to
the collection, which comes in at a higher price than even the Apple
HomePod. All bases are very much covered.
Google
knows that the hardware play is not a long-term solution. It is a
necessary strategy for the here and now, but Google will want to
convince other hardware producers to integrate the Assistant, much in
the same way it did with Android smartphone software. That removes the
expensive production costs but keeps the vital currency of consumer
attention spans.
This plan is already in action, with support just announced for a range of smart displays:
This
adds a new, visual element to consumer interactions with smart speakers
and, vitally, brings the potential to use Google Photos, Hangouts, and
YouTube.
Google
also wants to add a “more human touch” to its AI assistant and has
hired a team of comedians, video game designers, and empathy experts to
inject some personality.
Google
is, after all, an advertising company, so the next project will be to
monetize this technology. For now, the core aim is to provide a better,
more human experience than the competition and gain essential territory
in more households. The search giant will undoubtedly find novel ways to
make money from that situation.
Although
it was slower off the mark than Amazon, Google’s advertising nous and
growing range of products mean it is still a serious contender in both
the short- and long-term.
Apple
Hardware: Apple HomePod (Due to launch in 2018 at $349), iPhone, MacBooks, AirPods
Digital Assistant: Siri
Usage Statistics:
42.5% of smartphones have Apple’s Siri digital assistant installed (Highervisibility)
41.4 million monthly active users in the U.S. as of July 2017, down 15% on the previous year (Verto Analytics)
19% of iPhone users engage with Siri at least daily (HubSpot)
Overview:
Apple
retains an enviable position in the smartphone and laptop markets,
which has allowed it to integrate Siri with its OS in a manner that
other companies simply cannot replicate. Even Samsung, with its Bixby
assistant, cannot boast this level of synergy, as its smartphones
operate on Android and, as a result, have to compete with Google
Assistant for attention.
Nonetheless,
it is a little behind the curve when it comes to getting its hardware
into consumers’ home lives. The HomePod will, almost certainly, deliver a
much better audio experience than the Echo Dot or Google Home Mini,
with a $350 price tag to match. It will contain a host of impressive
features, including the ability to judge the surrounding space and
adjust the sound quality accordingly.
The
HomePod launch has been delayed, with industry insiders suggesting that
Siri is the cause. Apple’s walled garden approach to data has its
benefits for consumers, but it has its drawbacks when it comes to
technologies like voice recognition. Google has access to vast
quantities of information, which it processes in the cloud and uses to
improve the Assistant experience for all users. Apple does not possess
this valuable resource in anything like the same quantity, which has
slowed the development of Siri since its rise to fame.
That said, these seem likely to be short-term concerns.
Apple
will stay true to its core business strategy and it is one that is
served it rather well so far. The HomePod will sit at the premium end of
the market and will lean on Apple’s design heritage, with a focus on
providing a superior audio experience. It will launch with support for
Apple Music alone, so unless Apple opens up its approach to third
parties, it could be one for Apple fans only. Fortunately for Apple,
there are enough of those to ensure the product gains a foothold.
Whether its
Microsoft
Hardware: Harman/kardon Invoke speaker, Windows smartphones, Microsoft laptops
Digital Assistant: Cortana
Usage Statistics:
5.1% of smartphones have the Cortana assistant installed
Cortana now has 133 million monthly users (Tech Radar)
25% of Bing searches are by voice (Microsoft)
Overview:
Microsoft
has been comparatively quiet on the speech recognition front, but it
possesses many of the component parts required for a successful speech
recognition product.
With
a very significant share of the business market, the Office suite of
services, and popular products like Skype and LinkedIn, Microsoft
shouldn’t be written off.
Apple’s
decision to default to Google results over Bing on its Siri assistant
was a blow to Microsoft’s ambitions, but Bing can still be a competitive
advantage for Microsoft in this arena. Bing is a source of invaluable
data and has helped develop Cortana into a much more effective speech
recognition tool.
The
Invoke speaker, developed by Harman/kardon with Cortana integrated into
the product, has also been reduced to a more approachable $99.95.
There
are new Cortana-enabled speakers on the way, along with smart home
products like thermostats. This should see its levels of uptake
increase, but the persistent feeling is that Microsoft may be a little
late to this party already.
Where
Microsoft can compete very credibly is in the office environment, which
has also become a central consideration for Amazon. Microsoft is
prepared to take a different route to gain a foothold in this market,
but it could still be a very profitable one.
The Future of Speech Recognition Technology
We
are still some distance from realizing the true potential of speech
recognition technology. This applies both to the sophistication of the
technology itself and to its integration into our lives. The current
digital assistants can interpret speech very well, but they are not the
conversational interfaces that the technology providers want them to be.
Moreover, speech recognition remains limited to a small number of
products.
The rate of progress, compared to the earliest forays into speech recognition, is really quite phenomenal nonetheless.
As
such, we can look into the near future and envisage a vastly changed
way of interacting with the world around us. Amazon’s concept of
“ambient computing” seems quite fitting.
The
smart speaker market has significant room left to grow, with 75% of US
homes projected to have at least one by the end of 2020.
Now
that users are getting over the initial awkwardness of speaking to
their devices, the idea of telling Alexa to boil the kettle or make an
espresso does not seem so alien.
Voice is becoming an interface of its own, moving beyond the smartphone to the home and soon, to many other quotidian contexts.
We
should expect to see more complex input-output relationships as the
technology advances, too. Voice-voice relationships restrict the
potential of the response, but innovations like the Amazon Echo Show and
Google’s support for smart displays will open up a host of new
opportunities for engagement. Apple and Google will also incorporate
their AR and VR applications when the consumer appetite reaches the
required level.
Challenges
remain, however. First of all, voice search providers need to figure
out a way to provide choice through a medium that lends itself best to
short responses. Otherwise, how would it be possible to ensure that a
user is getting the best response to their query, rather than the
response with the highest ad budget behind it?
Modern
consumers are savvy and have access to almost endless information, so
any misjudgements from brands will be documented and shared with the
user’s network.
A
new study from Google has shown that there is an increasing acceptance
among consumers that brands will use smart speakers to communicate with
them. A sizeable number revealed a willingness to receive information
about deals and sales, with almost half wanting to receive personalized
tips:
Speech
recognition technology provides the platform for us to communicate
credibly, but it is up to marketers to make the relationship with their
audience mutually beneficial.
Key Takeaways
Brands
need to consider how they can make an interaction more valuable for a
consumer. The innate value proposition of voice search is that it is
quick, convenient, and helpful. It is only by assimilating with — and
adding to -this relationship between technology and consumer that they
will cut through. The Beauty and the Beast example provides an early,
cautionary tale for all of us.
Amazon
is in prime position to monetize its speech recognition technology, but
still faces obstacles. Sponsorship of Amazon’s Choice has been explored
as a route to gain revenue without losing customers.
Google
has made speech recognition a central focus for the growth of their
business. With a vast quantity of data at its disposal and increasing
third-party support, Google Assistant will provide a serious threat to
Amazon’s Alexa this year.
Marketers
should take advantage of technical best practices for voice search to
increase visibility today. While this technology is still developing, we
need to give it a helping hand as it completes its mammoth tasks.
The
best way to understand how people use speech recognition technology is
to engage with it frequently. Marketers serious about pinpointing areas
of opportunity should be conducting their own research at home, at work,
and on the go.
Have you read Paid Applications Agreement, Schedule 2, Section 3.8(b)?
If
you’ve ever submitted an app to the App Store, you know the frustration
when Apple rejects your submission. Even more so when you thought you’d
followed all the rules. As it turns out, Apple can bury requirements
wherever they want, and it’s your burden to keep up.
About
a year ago, Apple started rejecting apps that didn’t comply with
Schedule 2, Section 3.8(b) of the Paid Applications Agreement, a verbose
list of self-evident truths about subscriptions. The Paid Applications
Agreement is a 37-page document that you had to agree to before you
could submit your app. It is only available via iTunes Connect in the
form of downloadable PDF.
The actual contents of Schedule 2, Section 3.8(b):
I really like the part about privacy policies.
3.8(b)
requires that you “clearly and conspicuously disclose to users” all of
the above bullets. The first few items seem harmless enough but then we
start to get off into the weeds.
Apple
wants you to reproduce, “clearly and conspicuously”, all the details of
auto-renewing subscriptions. This information should be part of the
standard StoreKit subscription purchase flow. None of these bullets have
anything app specific to them. They are just boilerplate legalese.
iOS’s purchase UI, more than enough information.
Apple
has an iOS level user interface flow for in-app purchases that is quite
good as of iOS 11. This view already covers most of the in-the-weeds
bullets, except telling users about the 24-hour renewal policy.
Requiring
every developer to implement their version of 3.8(b) is costly and
creates a fractured experience for the user. Apple should be putting it
in the standard sheet. But it’s Apple’s walled garden. When they say
jump, you say “fine, whatever.”
How to Comply With 3.8(b)
According
to recent rejections that I’ve seen (as of Jan. 8th, 2018), reviewers
are being more particular about what your purchase flow requires. From a
recent rejection:
Adding
the above information to the StoreKit modal alert is not sufficient;
the information must also be displayed within the app itself, and it
must be displayed clearly and conspicuously during the purchase flow
without requiring additional action from the user, such as opening a
link.
All
of the information in 3.8(b) must be “displayed clearly and
conspicuously during the purchase flow without requiring additional
action from the user, such as opening a link.” Your beautiful and
compact purchase flow must include in it, somewhere, nine bullets
written by a lawyer.
Confide, recently updated, achieved it with the following:
According to one reviewer, being below the fold with a leading arrow qualifies as “clearly and conspicuously.”
For another data point, I know of one recently rejected developer who had the same information, but in another view that was linked from the purchase flow with a button. This did not qualify (according to one reviewer).
A Template
Include a customized version of the following “clearly and conspicuously” in your purchase flow:
A
[purchase amount and period] purchase will be applied to your iTunes
account [at the end of the trial or intro| on confirmation].
Subscriptions
will automatically renew unless canceled within 24-hours before the end
of the current period. You can cancel anytime with your iTunes account
settings. Any unused portion of a free trial will be forfeited if you
purchase a subscription.
For more information, see our [link to ToS] and [link to Privacy Policy].
Put
it on the screen where you initiate the in-app purchase, below the fold
might be OK, but you might want to put something to lead users there.
UPDATE:
Readers are telling me it may also be required that you include it in
your app store description. It’s a much easier change to include so I
recommend you add it there to.
Why has Apple Taken a Legal Problem and made it Ours?
Apple
shouldn’t be burying submission requirements in the bodies of contracts
that nobody will read. If Apple wants developers to know something,
they should put it in the App Store Guidelines, HIG, or developer
documentation. The cost of making changes in a software project right at
the end can be astronomical. Dropping a bomb like this on developers at
submission shows a total lack of regard for our costs.
Why
didn’t they just update the iOS in-app purchase sheet? I speculate that
Apple discovered some legal exposure from in-app subscriptions and
fixed it with lawyers instead of designers. This problem could be
universally solved with an iOS update, but I think some side effect of
Apple being a vast, lumbering bureaucracy made forcing 3.8(b) onto
developers the more politically convenient path. Apple, if you are
reading this, please either update the iOS sheet or move the
requirements to the App Store guidelines, so fewer developers get caught
unawares.
RevenueCat
is the best way to implement subscriptions in your mobile app. We
handle all the complicated parts so you can get back to building.
Request an invite today at https://www.revenuecat.com/
Hardik Gandhi is Master of Computer science,blogger,developer,SEO provider,Motivator and writes a Gujarati and Programming books and Advicer of career and all type of guidance.
