AT&T Labs Research — Leading Invention, Driving Innovation

AT&T unveils tool to help apps run faster, smarter

A free diagnostic tool is now available to help developers create energy-efficient apps that conserve battery power and respond faster.

The AT&T Application Resource Optimizer (ARO), which runs alongside device apps, analyzes energy usage and pinpoints inefficiencies in how apps connect and transfer data (a previous article details the complex device-network interactions). Go here to access ARO and learn more.

Commuting distances mapped using mobility data

Are mobile devices, by freeing people to work anywhere, affecting commuting distances?

To understand better, researchers from AT&T and MIT's Senseable City Lab used anonymous, aggregated cell phone data to depict commuting patterns in eight major US urban areas. The result of this research is a series of visualizations, currently featured in Esquire and at the MIT's Senseable City Lab site, that depict an expanding urban sprawl.

Improving on force-directed graph drawing

Using data from the Tree of Life project, Yifan Hu visualized genetic linkages among 93891 species using a high-performance graph layout method designed to overcome the peripheral warping sometimes seen with force-directed algorithms, especially when used on large data sets.

The result, shown here, is an aesthetic tree layout that efficiently utilizes white space to clarify the underlying phylogenetics structure.

Study looks at speaking rates and TTS intelligibility

AT&T researchers, along with the ASA Text-to-Speech (TTS) Technology working group (S3-WG91), are collecting data comparing the intelligibility of seven different synthetic speech systems at various speaking rates. The aim is to make TTS more usable by all, including users of mobile devices, children with learning disabilities, people with visual disabilities (see paper) or hearing impairments. 

To participate, click here or, if you are legally blind, here.

AT&T Labs - Research is hiring

AT&T Labs-Research is looking for talented individuals who can make a difference in computer science and electrical engineering research. Areas of interest include networking, wireless, information management and visualization, HCI and devices, statistics, machine learning, speech and language processing, software systems, cloud computing, and image processing. PhD-level candidates at all levels of experience may apply. Interviews will be conducted in early 2012. For more information, click here.

AT&T WATSON is coming to cars

The AT&T WATSON^SM speech and language engine is being incorporated into a prototype voice-enabled mobile virtual assistant for use by automotive manufacturers.

To help reduce driver distraction, AT&T is working with Panasonic Automotive and QNX Software to let drivers use voice commands for text messaging, web browsing, updating social media, and other tasks, including controlling GSP and radio. See the press release for more information.