Babies’ babbling enables new research tool

Thirty two years ago, com­puter and infor­ma­tion sci­ence pro­fessor Har­riet Fell had just given birth to her oldest child when Linda Fer­rier, then a PhD stu­dent working on her thesis in speech lan­guage pathology at Boston Children’s Hos­pital came into her hos­pital room looking for infant vol­un­teers. She was col­lecting record­ings of infants “bab­bling” to better under­stand how they devel­oped the capacity to pro­duce speech over time. A young researcher her­self, Fell thought this idea was great and signed her daughter up. For the next two and a half years, Fer­rier vis­ited their home each month to cap­ture hour-​​long record­ings of the young child cooing and chat­ting as babies do. After the thesis was com­plete, Fer­rier receded into the family’s memory, until one day, when Fell was drop­ping Tova off at the North­eastern day care center. Oppo­site the day care she saw Ferrier’s name freshly minted on an office door. The doc­toral stu­dent was now a pro­fessor at Fell’s own institution.

The first project Harriet Fell and Linda Ferrier collaborated on was called the Baby Babble Blanket, which allowed infants to produce sounds—a mother's voice, another baby babbling, a toilet flushing—by rolling around and pushing buttons on the blanket. Photo from Thinkstock.

The first project Har­riet Fell and Linda Fer­rier col­lab­o­rated on was called the Baby Babble Blanket, which allowed infants to pro­duce sounds—a mother’s voice, another baby bab­bling, a toilet flushing—by rolling around and pushing but­tons on the blanket. Photo from Thinkstock.

The two imme­di­ately struck up a working rela­tion­ship, coming up with cool projects for their senior cap­stone stu­dents to col­lab­o­rate over. “It was a lot of fun,” said Fell. “My stu­dents would build things, her stu­dents would test things, and every­body would make sug­ges­tions.” Over time the edu­ca­tional col­lab­o­ra­tion blos­somed into a research part­ner­ship, with Fell designing com­plex soft­ware for novel speech research tools Ferrier—now retired—wanted to try out. Their first project together was called the Baby Babble Blanket, which allowed infants to pro­duce sounds—a mother’s voice, another baby bab­bling, a toilet flushing—by rolling around and pushing but­tons on the blanket. The tool was designed for infants with motor and neu­ro­log­ical dis­or­ders to, as the researchers put it, “estab­lish cause and effect skills, explore a bab­bling reper­toire like normal infants, and use early motor move­ments to pro­duce dig­i­tized sounds.” One child, Fell recalled, was par­tic­u­larly fond of the toilet flushing sound and would con­tin­u­ally bop his head on the part of the blanket that pro­duced that noise.

After this, Fer­rier and Fell con­tinued to explore ways they could use com­puters to explore ques­tions about speech. “My interest was trying to detect med­ical prob­lems in their speech or sounds,” said Fell. “I just had the feeling that you could tell cer­tain things about babies that could not easily be rec­og­nized. At the time there had been research on neonatal cry, where cer­tain fea­tures  could be rec­og­nized that indi­cate neu­ro­log­ical prob­lems. I thought that if the artic­u­la­tors aren’t working right, or maybe the parts of the brain that con­trol speech, there’s a cer­tain ten­sion or dis­tur­bance, then maybe you could just tell this from the acoustic signal.”

She devel­oped a soft­ware pro­gram that ana­lyzes the struc­ture of syl­la­bles, looking for variety, dura­tion, that kind of thing, in the sounds people create with their mouths. It could be used for adults just as well as with infants, she said. In one project, a col­league used the soft­ware to detect fatigue in adult speech. Fer­rier used it to work as an accent reduc­tion tool with for­eign lan­guage speakers.

Fell soon came to realize that what she had orig­i­nally devel­oped as a tool for very spe­cific research ques­tions could actu­ally be quite useful for a variety of inves­ti­ga­tions. But back when this was all get­ting off the ground, com­puters were still the size of a broom closet and learning the soft­ware had a very high learning curve. In the last few decades, of course, the per­sonal com­puter has pretty much taken over the world, so the con­straints that once kept Fell from releasing her pro­gram to the masses have now van­ished. About 15 years ago, she joined forces with an entre­pre­neur named Joel MacAuslan. A year ago they applied for and received a Small Busi­ness Inno­va­tion Research grant from the National Insti­tutes of Health to develop the soft­ware they had been using for their own studies into some­thing more use­able and user friendly.

Now a group of alpha testers is using the soft­ware to explore a variety of speech lan­guage ques­tions, looking at every­thing from early vocal sig­na­tures of autism to the speed at which adults are able to cog­ni­tively respond to audi­tory stimuli.

Before our users were infants,” Fell said. “Now they are sci­en­tists.” What began as a col­lab­o­ra­tion between friends has clearly grown into a research tool that has the poten­tial to enable a host of new inves­ti­ga­tions. The only limit now is the sci­en­tists’ imaginations.