Purdue professor partners with Facebook to develop new way of communicating

A Purdue University professor who set out to help blind and deaf people better communicate has in the process developed a system that could make it possible for everyone to receive voice, text or email messages without having to pick-up or even look at their cell phones or computers.

Hong Tan, a professor of electrical and computer engineering, was the lead professor on the research that developed a method of receiving messages by learning to interpret signals such as a buzzing sensation through the skin on the forearm.

Hong Tan

The year-long project was done in conjunction with researchers from the Massachusetts Institute of Technology and Facebook. 

The research was funded by Facebook and the findings were presented June 15 at the Proceedings of EuroHaptics 2018 conference in Pisa, Italy. Purdue officials gave a hands-on demonstration of the system at the conference on June 16.

Purdue researchers are collaborating with Facebook through the California-based company’s Sponsored Academic Research Agreement.

“I’m excited about this," said Tan, founder and director of Purdue’s Haptic Interface Research Laboratory. "Imagine a future where you’re able to wear a sleeve that discreetly sends messages to you, through your skin in times when it may be inconvenient to look at a text message,”

“When that happens, the hearing-impaired, the visually-impaired, everyone can benefit," she said. "I’m really hoping this takes off as a general idea for a new way to communicate."

Facebook, Tan said, is interested in developing new platforms for communication and touch-related research Purdue is doing has been promising.

Tan told IBJ that she’s “not aware of any particular commercial development plans” for her research at this time. The terms of the contract with Facebook, including stipulations regarding intellectual property and Facebook’s level of funding, are part of a non-disclosure agreement, she explained. 

“We will, however, be publishing the results of our experiments in the coming months, so please stay tuned,” Tan said. “As for my group, we are seeking research funding from other sources to continue this work.”

In the study, subjects used a cuff—not unlike the one used to measure blood pressure—encircling the forearm from the wrist to below the elbow. The instrument, wrapped around the test subject’s non-dominant arm, featured 24 sensors that, when stimulated, emit a vibration against the skin, changing quality and position in the process.

Tan said the 39 phonemes (units of sound in a language that distinguish one word from another) in the English language were mapped and could be sent to the sensors on the cuff. The sounds of consonants were stationary sensations on different areas of the arm while vowels were indicated by stimulations that moved up, down or around the forearm.

The Purdue-led research showed that a phonemes-based system worked better than a word-based approach in terms of helping users learn the system more quickly.

While it may sound complicated, Tan said the system is relatively easy to learn. Many people tested in the system demonstrated 80 percent proficiency with a 100-word vocabulary with 1 hour and 40 minutes of training. The training took place in 10-minute intervals.

“For this research, the learning progress is one of the key things,” Tan said. “With this, not only do we have a system that works, but we’re able to train people within hours rather than months or even years.”