For anyone who has followed CES 2020, one of the announcements that created the most buzz was Samsung's NEONs (neo-humans), or AI assistants that resemble humans. These NEONs are extremely lifelike--so much so that when you look at them, it can be hard to believe they aren't real people on the other end of the video. But will lifelike AI assistants really be the future? What practical use will they have if they are? And will these AI assistants find their way into the enterprise or just become nothing more than a consumer-focused gimmick?

Word embeddings are a powerful approach to capturing semantic similarity among terms in a vocabulary. In this paper, we develop exponential family embeddings, which extends the idea of word embeddings to other types of high-dimensional data. As examples, we studied several types of data: neural data with real-valued observations, count data from a market basket analysis, and ratings data from a movie recommendation system. The main idea is that each observation is modeled conditioned on a set of latent embeddings and other observations, called the context, where the way the context is defined depends on the problem. In language the context is the surrounding words; in neuroscience the context is close-by neurons; in market basket data the context is other items in the shopping cart.

Hybrid methods that utilize both content and rating information are commonly used in many recommender systems. However, most of them use either handcrafted features or the bag-of-words representation as a surrogate for the content information but they are neither effective nor natural enough. To address this problem, we develop a collaborative recurrent autoencoder (CRAE) which is a denoising recurrent autoencoder (DRAE) that models the generation of content sequences in the collaborative filtering (CF) setting. To do this, we first develop a hierarchical Bayesian model for the DRAE and then generalize it to the CF setting. The synergy between denoising and CF enables CRAE to make accurate recommendations while learning to fill in the blanks in sequences.

Low-rank matrix approximation (LRMA) methods have achieved excellent accuracy among today's collaborative filtering (CF) methods. In existing LRMA methods, the rank of user/item feature matrices is typically fixed, i.e., the same rank is adopted to describe all users/items. However, our studies show that submatrices with different ranks could coexist in the same user-item rating matrix, so that approximations with fixed ranks cannot perfectly describe the internal structures of the rating matrix, therefore leading to inferior recommendation accuracy. In this paper, a mixture-rank matrix approximation (MRMA) method is proposed, in which user-item ratings can be characterized by a mixture of LRMA models with different ranks. Meanwhile, a learning algorithm capitalizing on iterated condition modes is proposed to tackle the non-convex optimization problem pertaining to MRMA. Experimental studies on MovieLens and Netflix datasets demonstrate that MRMA can outperform six state-of-the-art LRMA-based CF methods in terms of recommendation accuracy.

Speaker recognition systems based on deep speaker embeddings have achieved significant performance in controlled conditions according to the results obtained for early NIST SRE (Speaker Recognition Evaluation) datasets. From the practical point of view, taking into account the increased interest in virtual assistants (such as Amazon Alexa, Google Home, AppleSiri, etc.), speaker verification on short utterances in uncontrolled noisy environment conditions is one of the most challenging and highly demanded tasks. This paper presents approaches aimed to achieve two goals: a) improve the quality of far-field speaker verification systems in the presence of environmental noise, reverberation and b) reduce the system qualitydegradation for short utterances. For these purposes, we considered deep neural network architectures based on TDNN (TimeDelay Neural Network) and ResNet (Residual Neural Network) blocks. We experimented with state-of-the-art embedding extractors and their training procedures. Obtained results confirm that ResNet architectures outperform the standard x-vector approach in terms of speaker verification quality for both long-duration and short-duration utterances. We also investigate the impact of speech activity detector, different scoring models, adaptation and score normalization techniques. The experimental results are presented for publicly available data and verification protocols for the VoxCeleb1, VoxCeleb2, and VOiCES datasets.

It seems like just about everything works with Alexa in some form or another, and the Echo Wall Clock is no exception. This timekeeper is Amazon's latest attempt to make everyday products easier to use--and it kind of succeeds at doing that. When using the clock, you can set multiple timers and reminders, and, instead of using your Amazon Echo speaker to keep track of how much time is left, you can see how much time remains when you look at the analog clock. That's because the Echo Wall Clock, which retails on Amazon for $29.99, has 60 LED tick marks that light up to show you how much time is left. It takes several minutes to get the Echo Wall Clock up and running, and you'll need an Echo speaker nearby to do so.

Amazon's first attempt at a set of true wireless earbuds gets a lot right, with Bose active noise reduction technology and hands-free Alexa. At £119.99, the Echo Buds undercut rivals, some of which cost more than twice as much. Their design is generic: large, kidney-shaped with a glossy touch panel on the outside and a standard silicone eartip on the inside. The eartip supports the earbud with the majority of the rest of the body sitting outside the ear. But the earbuds are large and heavy at 7.6g each, meaning they sit proud of your ear.

In 2019, smart speaker ownership in the U.S. surpassed 76 million, according to CIRP (Consumer Intelligence Research Partners), up from 66 million at the end of 2018. As voice-assistant technologies continue to advance in smartphones, smart speakers, and in vehicles, more consumers and businesses are exploring the benefits of using this technology to make tasks simpler. Tech giants like Amazon, Apple, Google, and Microsoft are all chomping at the bit, hoping customers will turn to their voice assistants for increasingly complex tasks. CIRP's research suggests Amazon is the current leader in smart speakers, with about 70% of the installed base being Amazon Echo devices, 25% being made up of Google Home devices, and 5% being made up of Apple HomePods. This remains relatively unchanged from the firm's previous year's research, with the exception of slight percentage growth from both Google and Apple in 2019 compared to 2018.

As Valentine's Day approaches and the aroma of love turns even devout singletons into frenzied love-seekers, many will invariably turn to dating apps for help. But caving in and venturing into the murky world of Hinge, Tinder and Bumble is a poisoned chalice, doomed to fail even if it works, a new study reveals. Academics have found people who have success in the fickle world of virtual swiping perceive themselves to be desirable as a result of their conquests. This sense of self-desirability, it has been proved, makes a person more likely to cheat when they eventually settle down into a serious relationship. Dr Cassandra Alexopoulos of the University of Massachusetts led the research and quizzed 395 participants on their dating app use.

Fox News Flash top entertainment and celebrity headlines are here. Check out what's clicking today in entertainment. Is Ben Affleck on a dating app or not? According to Patti Stanger, he is. The "Millionaire Matchmaker" star recently appeared on Us Weekly's "Hot Hollywood" podcast, where she claimed she's seen the 47-year-old actor on the elite dating app, Raya, on multiple occasions.