Amazon studies user voice recordings to make improvements to Alexa, according to the company. Each time you say "Alexa" (or whatever your wake word of choice happens to be), Amazon might be listening. However, the company says the wake word and the command or question that follows is the only part that's being recorded. This is all to help Alexa get smarter, but this may leave some smart home users feeling a bit uneasy. While Amazon points out that an "extremely small fraction of voice recordings are manually reviewed," the recordings have been mismanaged in the past, including a few horror stories, so it doesn't hurt to take precautions. You can adjust Alexa's voice recording settings in the Amazon Alexa app, available for download on iOS and Android devices. There are other ways to manage your voice recordings in the Amazon Alexa app, like automatic deletion and turning off voice recordings altogether.

Paolo Pirjanian is an Armenia born in Iran and fled to Denmark as a teen. From the time he was young, he was fascinated by computers and started coding in his bedroom. After getting his PhD in robotics, Paolo became an early leader in the field of consumer robotics who has 16 years of experience developing and commercializing cutting-edge home robots. He worked at NASA JPL and led world-class teams and companies at iRobot, Evolution Robotics, and others. In 2016, Paolo founded Embodied, Inc. with the vision to build socially and emotionally intelligent digital companions that improve care and wellness and support people in living better lives every day.

"Alexa, play Beyoncé!" Ever wondered how this works? Want to find out about the technology behind virtual personal assistants and other smart home gadgets? Then this course is for you! Performing via cognition, today's computers, software and devices can recognize people and speech, as well as write, speak and read handwriting. This course is perfect for teens with previous Python experience who want to develop a practical understanding of ML.

One of the biggest tech opportunities over the next decade is artificial intelligence (AI). According to Grand View Research, the general AI market, across a variety of products and services, is set to expand at a huge 46.2% average annual growth rate between 2019 and 2025. AI is set to power the next generation of tech innovation, from virtual assistants, to big-data analytics and inference, to self-driving cars, to image and voice recognition, and new 5G applications. With COVID-19 sweeping across the country and leading to shelter-in-place orders, many of these tech trends could be accelerated, as business look to quickly digitize to serve customers and cut costs. With many AI-oriented big tech companies reporting resilient financial results in their first quarter earnings reports, several look like strong buys in May as we come out of the March crash.

When unexpected events occur and customer demand increases what can companies do to smooth out the resulting peaks and troughs? Henry Jinman at EBI.AI outlines four ways conversational AI can help and shares a real-life example. First there were Storms Ciara and Dennis, now Coronavirus. The impacts of the pandemic are changing every minute, causing widespread fear and uncertainty around the world. Frontline teams are under particular pressure as agents struggle to manage surging and often widely fluctuating volumes of customer calls. When disaster strikes, what can businesses do to smooth the peaks and troughs?

You've figured out the basics of setting up your smart home, now it's time to raise your game. I've spent years installing, configuring, and tweaking dozens of smart home products in virtually every product category. Along the way I've figured out a lot of the secrets they don't tell you in the manual or the FAQs, ranging from modest suggestions that can make your smart home configuration less complex, to essential decisions that can save you from having to start over from scratch a few years later. Here's my best advice on how to optimize your smart home, digested into a dozen top tips and best practices. These days, an Amazon or Google/Nest smart speaker or smart display can fill the role of a smart home hub (and some Amazon Echo devices are equipped with Zigbee radios).. There are three major smart home platforms on the market, and your smart home will probably have at least one of them installed: Amazon Alexa, Google Assistant, or Apple HomeKit.

Sonos unveiled three all-new higher-end speakers today: The $799 Sonos Arc, a roundish soundbar with support for Dolby Atmos object-based audio; the $699 Sonos Sub (Gen 3), a dual-driver, dual-amplifier subwoofer; and the $499 Play 5, a 14-pound speaker with six drivers and an equal number of Class D amplifiers. All three new speakers will be available in all-white or all-black (including the Sonos logo) on June 10, preceded by the release of the new Sonos S2 app on June 8. The Sonos Arc replaces both the Sonos Playbar, the company's first soundbar that came to market in 2013, and the Sonos Playbase that shipped in 2018. The less-expensive Sonos Beam ($399) will remain in the company's lineup. As with several newer Sonos speakers, the Arc supports your choice of Amazon Alexa or Google Assistant along with Apple AirPlay 2. The roundish Sonos Arc can operate on furniture or be hung on the wall using an optional mounting kit.

When it comes to headphones, especially truly wireless versions, it's hard to top those made by a certain Cupertino-based company named after a popular fruit. Google's own attempt to recreate the magic that is the Apple AirPods is here, and it's safe to say the reboot of the Android-friendly Pixel Buds is a pretty good one. The $179 truly wireless earbuds ditch the connecting wire found on the previous model, but maintain all the features that made the last version so appealing. While issues like fit and minor quirks -- especially with the AI-powered Google Assistant -- still exist, the end result is a pair of truly wireless buds with the same high-end build quality as the competition, and a welcome addition to the Android ecosystem. If only they didn't chafe so much.

Tinder is set to make online dating amid coronavirus lockdowns more personal. The dating site is adding a video-chatting feature to its popular app by the end of June, allowing those in isolation to virtually meet one-on-one. The feature was announced during the Match Group's, Tinder's parent company, first quarter earnings call and this is the largest addition to the platform since launching in 2012. Along with the video chatting service, Tinder also shared that it has seen a surge in usage during the pandemic - with women users swiping 37 percent more each day than before. Tinder is set to make online dating amid coronavirus lockdowns more personal.

To build Sounding Board, we develop a system architecture that is capable of accommodating dialog strategies that we designed for socialbot conversations. The architecture consists of a multi-dimensional language understanding module for analyzing user utterances, a hierarchical dialog management framework for dialog context tracking and complex dialog control, and a language generation process that realizes the response plan and makes adjustments for speech synthesis. Additionally, we construct a new knowledge base to power the socialbot by collecting social chat content from a variety of sources. An important contribution of the system is the synergy between the knowledge base and the dialog management, i.e., the use of a graph structure to organize the knowledge base that makes dialog control very efficient in bringing related content to the discussion. Using the data collected from Sounding Board during the competition, we carry out in-depth analyses of socialbot conversations and user ratings which provide valuable insights in evaluation methods for socialbots. We additionally investigate a new approach for system evaluation and diagnosis that allows scoring individual dialog segments in the conversation. Finally, observing that socialbots suffer from the issue of shallow conversations about topics associated with unstructured data, we study the problem of enabling extended socialbot conversations grounded on a document. To bring together machine reading and dialog control techniques, a graph-based document representation is proposed, together with methods for automatically constructing the graph. Using the graph-based representation, dialog control can be carried out by retrieving nodes or moving along edges in the graph. To illustrate the usage, a mixed-initiative dialog strategy is designed for socialbot conversations on news articles.