Voice assistants are becoming an essential part of our daily lives. When Apple's Siri hit markets in 2011, it managed to gain an impressive attraction of tech enthusiasts, yet no one was certain about how this novelty shall bring a tech revolution. Today, we are regular users of Google Voice Assistant, Amazon Alexa, and many more. Things took a turn when Google Home, Amazon Echo, and Apple HomePod went mainstream in 2017. All these instances converge on how voice assistants are proving themselves as a tech enabler with impressive possibilities. Not only in households, but they are also slowly proving to be useful in the business quarters too.

Amazon's Echo line is the reigning champ of the smart speaker world. Compared to virtual assistants like Google and Siri, Alexa works with far more gadgets and responds to significantly more commands than the competition. Plus, Echo devices are pretty inexpensive, starting around $25 for the Flex and $30 for the Dot when you catch it on sale. I bought six of these Echo Shows for Christmas presents. They have a nice screen, so all the kids and grandkids can pop in and say hello to my Mom whenever they want.

In today's world, it is nearly impossible to avoid voice-controlled digital assistants. From the interactive intelligent agents used by corporations, government agencies, and even personal devices, automated speech recognition (ASR) systems, combined with machine learning (ML) technology, increasingly are being used as an input modality that allows humans to interact with machines, ostensibly via the most common and simplest way possible: by speaking in a natural, conversational voice. Yet as a study published in May 2020 by researchers from Stanford University indicated, the accuracy level of ASR systems from Google, Facebook, Microsoft, and others vary widely depending on the speaker's race. While this study only focused on the differing accuracy levels for a small sample of African American and white speakers, it points to a larger concern about ASR accuracy and phonological awareness, including the ability to discern and understand accents, tonalities, rhythmic variations, and speech patterns that may differ from the voices used to initially train voice-activated chatbots, virtual assistants, and other voice-enabled systems. The Stanford study, which was published in the journal Proceedings of the National Academy of Sciences, measured the error rates of ASR technology from Amazon, Apple, Google, IBM, and Microsoft, by comparing the system's performance in understanding identical phrases (taken from pre-recorded interviews across two datasets) spoken by 73 black and 42 white speakers, then comparing the average word error rate (WER) for black and white speakers.

Tinder, the most popular dating app in the world, has banned teens under the age of 18 but it's not stopping them from signing up. A Massachusetts man is accused of kidnapping and assaulting a woman he met on Tinder, threatening to kill her and her child if she went to the cops, authorities said. Peter Bozier, 28, was arrested Tuesday during a traffic stop in Sudbury after the victim told investigators she was severely beaten and strangled while being held against her will at Bozier's home, police said. The victim said the harrowing ordeal began a day earlier, police spokesman Lt. Robert Grady told the MetroWest Daily News. Grady said the woman managed to "release herself from the situation" and then went to a hospital in Burlington, where hospital staffers contacted police, the newspaper reported.

AI assistants have started carrying out tasks on a user's behalf by interacting directly with the web. However, training an interface that maps natural language (NL) commands to web actions is challenging for existing semantic parsing approaches due to the variable and unknown set of actions that characterize websites. We propose FLIN, a natural language interface for web navigation that maps NL commands to concept-level actions rather than low-level UI interactions, thus being able to flexibly adapt to different websites and handle their transient nature. We frame this as a ranking problem where, given a user command and a webpage, FLIN learns to score the most appropriate navigation instruction (involving action and parameter values). To train and evaluate FLIN, we collect a dataset using nine popular websites from three different domains. Quantitative results show that FLIN is capable of adapting to new websites in a given domain.

Only 10 years ago, artificial intelligence (AI) was just a lofty concept for consumers, appearing in pop culture references or fleeting news stories. Today, it pervades every corner of life, from Siri on our iPhones, to smart home security systems, to the recommended products in our Amazon feed. Everywhere we look, AI has become part of our daily processes -- and as we live, learn and work from home amidst the pandemic, this has only accelerated. It's safe to say that AI is no longer just a novel concept; it's a convenience we've come to expect in day-to-day life. What's interesting to me is that, in customer care, the benefits of AI are not quite so widely welcomed.

Amazon's smallest Echo has evolved quite a bit over the years. The first Amazon Echo Dot was small and puck-like but didn't have very good audio. In 2018, the company upgraded the Dot's speakers to a new 1.6-inch driver that gave it a lot more bass and overall better performance, plus it had a much more stylish fabric-clad exterior. Last year, Amazon added a new model called the Echo Dot with Clock, which is basically the same thing but with a digital clock on the front. In 2020, however, the company has decided to go… round.

Amazon users in the UK can now try and answer questions that Alexa doesn't know. The US tech company has announced the general availability of Alexa Answers in the UK – a crowd-sourced method of making its Alexa digital assistant more intelligent. The online hub offers users the chance to answer questions that Amazon's smart assistant Alexa didn't know the answer to. Users just need to sign in to their Amazon account at the Alexa Answers webpage and start browsing unanswered questions that they think they can answer. The UK launch will help Alexa get smart on topics specific to the UK, including the Spice Girls and the two-pound coin, Amazon hopes. In return for their knowledge, Alexa Answers users can earn points and get onto leaderboards on the hub.

The smart speaker that got things started is back, and it looks a little different now. Nathan Ingraham reviewed the new spherical Amazon Echo, and the good news is that no matter what you think of its looks, it sounds better than ever. Adding an extra tweeter -- not to mention the built-in Zigbee home hub -- seems to have made all the difference. That odd shape does mean its indicator light is a bit hidden, but when the sound is good enough that buying two to create a stereo setup seems like a reasonable option, maybe we can get over it… maybe. Garmin is offering Twitch broadcasters and other game streamers a way to layer their heart rate and other metrics into their streams, with an Esports Edition of its Instinct GPS smartwatch.

This paper proposes a theoretical analysis of recommendation systems in an online setting, where items are sequentially recommended to users over time. In each round, a user, randomly picked from a population of $m$ users, requests a recommendation. The decision-maker observes the user and selects an item from a catalogue of $n$ items. Importantly, an item cannot be recommended twice to the same user. The probabilities that a user likes each item are unknown. The performance of the recommendation algorithm is captured through its regret, considering as a reference an Oracle algorithm aware of these probabilities. We investigate various structural assumptions on these probabilities: we derive for each structure regret lower bounds, and devise algorithms achieving these limits. Interestingly, our analysis reveals the relative weights of the different components of regret: the component due to the constraint of not presenting the same item twice to the same user, that due to learning the chances users like items, and finally that arising when learning the underlying structure.