Ron Scherer (right) discusses "M5" experiments with Inoka Nanayakkara (left), a master's degree student doing research on vocal fold anatomy, and Nandhakumar Radhakrishnan (center), a doctoral student researching Hindustani singing.

Scherer, collaborators seek answers to mysteries of voice

How does airflow through the human larynx become sound? How are the wide ranges of pitch and loudness in our voices created, and when does a voice sound “natural”?

The questions may seem basic, but the answers are still unknown. Finding them is among the goals of a long-term research project led by Ronald Scherer, communication disorders.

Scherer and collaborators from Purdue University and the universities of Toledo and Cincinnati are in the third year of a second, four-year grant from the National Institutes of Health to study phonation using aerodynamic and acoustic models. The $2.5 million in grant funding is divided among the four participating universities.

Phonation refers to the vibration of the vocal folds in the larynx that modulates the air from the lungs, creating sound. The two vocal folds move back and forth very quickly as air passes, and the changing airflow creates sound, but how that happens remains a mystery, Scherer says.

He and his fellow researchers are trying to solve it with the help of different models. They include computer animation—also involving Comer Duncan and Lewis Fulcher, professors of physics and astronomy at BGSU, and their students—as well as “M5.”

M5 is the fifth in a series of models constructed to measure airflows and pressures in a simulated larynx. This one, Scherer says, is the most sophisticated model of its kind in the world, a Plexiglas larynx enlarged seven and a half times with pairs of “vocal folds” frozen in a snapshot of phonation.

Air is pulled through the model, and pressure is measured at various taps so researchers know the pressure on a fold at each point of its vibration. Measurements are taken with differing angles between the folds—nine different pairs can be inserted to cover the range of phonation from 40 degrees convergent to 40 degrees divergent—and varying distances between them. Those distances are governed by shims inserted on the sides of the “larynx.”

Recorded air pressures are used in computer models of phonation, whose dynamics must be studied to see how the process works normally, Scherer says. And empirical, physical modeling is imperative for measuring those pressures, he notes, saying that the project has compiled the most complete collection of laryngeal pressures found anywhere.

Unlike researchers who have used incorrect equations in the past, “we are using empirical pressures to drive the computer models,” according to the project director. Calling it a benchmark pressure distribution model that others can also use, he adds, “I’d say that’s the most important progress so far.”

Scherer also lauds a finding by research assistant Meena Agarwal about the false vocal folds, which are located above the true folds. In research for her doctoral degree, Agarwal found that the false folds, in a certain position, may help the true folds with phonation, reducing resistance to airflow and helping the true folds’ vibration to produce a louder sound.

“It’s an amazing finding,” although further modeling work remains to be done, says Scherer.

To his knowledge, Agarwal’s research is the only systematic study of the effects of the false vocal folds. Her analysis of the false folds in singers has indicated that they may aid singing if somewhat closer together. But if they’re too close, the false folds can produce a low-pitched, rough voice that can be a sign of a problem, Scherer points out.

Potential applications for performance and pathology can be found in the larger project as well.

“How come some people have big, bright voices? Are they creating sounds that other people don’t?” he asks. The answers, he says, have to do with laryngeal flow right above the glottis (the opening between the vocal folds). Scherer adds, too, that aspects of auditory perception have to be considered along with phonation when discussing voice quality.

Because problems with laryngeal tissue—including nodules, polyps, cancer and even paralysis of a vocal fold—will redistribute airflow and change pressures, the research needs to model such pathological tissue, he maintains. “All these conditions are challenges relative to modeling. A goal of the grant is to meet these challenges,” Scherer says, expressing hope that funding will be renewed past 2006.

The researchers, he continues, would also like their study to answer the question of how individual surgical procedures will affect the voice. Why else do the work, he asks, if not to help such professionals as surgeons, singing coaches and speech pathologists.

“We need to help solve the practitioners’ problems,” says Scherer, who works with voice students in music and is a consultant to supervisors in BGSU’s Speech and Hearing Clinic. “It’s from research and the needs of the clinic and teachers of voice, whether it’s singing or acting, that tell us what our research agenda should be.”