When it comes to critics of machine-generated speech, three-year-olds might be the toughest around. If they are going to sit and listen to their favorite stories, a witch better sound evil and a wolf better huff and puff.
Current text-to-speech (TTS) systems are not up to the challenge of expressing the often strong emotions in children’s stories. AT&T researchers are looking to add expression for dramatic effect with the StorEBook project, which uses word prominence to not only please the young set but also to improve TTS for everyone. Read more
Digital services in the car have the potential to make driving easier and safer—if they don’t distract the driver. A haptic steering wheel, by using tactile-based signals, may more safely convey navigation information than audio or visual cues.
Other cloud-based services recently showcased by AT&T Research include a bio-acoustic door lock, a car system that alerts when you forget an item, and Air Graffiti for air-tagging locations. Read more.
Spam texts are a growing problem. They add charges to some bills, attempt to trick users into divulging private information, and hog cellular bandwidth, slowing service. To combat spam texts, AT&T with other telecoms made the short code 7726 available for reporting spam.
AT&T is doing more. Using knowledge of the network and spamming patterns, researchers are working to identify and shut down spammers even before the first customer reports. Read more
With hundreds of channels, how do you find what you want to watch?
It's easy if you’re an AT&T U-verse® TV customer because you can now use the Easy Remote app to turn your iPhone or iPad into a voice-controlled remote. Just say what you want, and Easy Remote will return a list of shows matching your search criteria. Easy Remote relies on AT&T WATSONSM and other AT&T Research technologies, and was productized with help from the AT&T Foundry. Read more
Last year, AT&T released the AT&T Application Resource Optimizer (ARO) as a diagnostic tool so developers could optimize their existing apps for the cellular network. This year, AT&T is releasing AT&T ARO as open-sourced software. Now developers can incorporate ARO into their app development environment, ensuring apps that right from the beginning conserve battery energy and respond quickly to user input. Read more
A paper co-authored by Divesh Srivastava, Dense Subgraph Maintenance under Streaming Edge Weight Updates for Real-time Story Identification, has won the Best Paper Award for VLDB 2012.
The paper describes a real-time technique for identifying trending news stories within social media, an enormously rich information stream. The award cites the technique’s insightful algorithmic contributions and its potential for broader applicability.
This year’s ACM SIGMETRICS conference conferred its Test of Time Award on Network Tomography on General Topologies, a paper co-authored by Nick Duffield. The award, given each year to a previous conference paper whose impact is still felt 10-12 years after publication, cited the paper’s pioneering work in network tomography, particularly its novel methods to estimate delay, packet loss, and other measures of performance.
Profiled on the AT&T Labs website, Subhabrata (Shubo) Sen talks about his early interest in math and mechanics, and his current work on the hard problems of network management. As a part of his research, he has been working on improving user experience and network efficiency. This work has led to the AT&T Application Resource Optimizer (ARO). As described in another article on this page, ARO helps developers make mobile apps more battery-friendly and more responsive to user input.
Kevin Li as an AT&T researcher has the freedom to work on projects that aren’t just six months out. Given this opportunity, he is studying how people naturally interact with the world and using this information to design interfaces that require less cognitive effort to learn. One project, which he describes in this profile, is a haptic-enabled steering wheel that delivers navigation information using tactile signals, rather than visual or audio cues that can distract drivers and interfere with safe driving.
Lana Yarosh wants to make positive changes in people's lives by making communications technologies more engaging and more fun for families. New to AT&T Labs - Research, she has the requisite expertise in computer science, but she also has a background in design and psychology, giving her more insight into understanding people and what they want out of technology. In this profile, she talks about how her background helps her as an AT&T researcher.
Inspired by the engineers in his family and by the study of math and science, Emiliano Miluzzo wants to make a big impact on the lives of millions of people.
He may get his chance. As an AT&T researcher focused on mobile devices and pervasive computing, he’s using the latest machine-learning and interface techniques to create entirely new ways for devices and people to interact. As he says in this profile, the goal is to improve daily life in many dimensions.
Taniya Mishra’s work and life revolve around language and, now, children. A speech researcher who speaks five languages, she works on expressive speech synthesis (see profile). Her StorEBook project, a children's e-reader, explores how best to make voices embody a character’s emotions so a wolf sounds scary and a teacup sounds cute. Her own children (serving as a built-in user base) love the "story game" though probably miss its larger goal—improving speech synthesis for everyone.
In a profile on the AT&T Labs website, Oliver Spatscheck talks about what inspired his interest in computers as well as his research today exploring the complex interactions between network layers. His work in measuring these interactions and understanding them in depth helped lead to the AT&T Application Resource Optimization (ARO), an open-source diagnostic tool to help developers optimize their apps for better performance and longer battery life.
Parallel hybrid heuristics for the permutation flow shop problem
Mauricio Resende, Fed. U. of Minas Gerais Martin G. Ravetti, U. of Newcastle Carlos Riveros, U. of Newcastle Alexandre Mendes, Panos M. Pardalos