In 1995, the Schein Voice and Laryngeal Center of the Department of Otolaryngology at the Long Island Jewish Medical Center in New York established The Voice Wellness Program, a privately-funded, preventative voice healthcare program incorporating educational outreach and individual examination that encourages young singers and actors to obtain baseline vocal function evaluations. The program, provided at no charge, is designed to promote healthy voice use and longevity of professional career.
Recently, the Voice Wellness Program expanded to include the Lindemann Young Artist Development Program (YADP) of the Metropolitan Opera, and embarked upon a research data-gathering documentation project designed to quantify some of the measurable aspects of the singing voice. The long-term goal of this research component of the Voice Wellness Program is to obtain data that increase general understanding of the singing voice and may ultimately be used by physicians, vocal pedagogues and singers themselves towards attaining maximum vocal potential and longevity of performance career.
In pursuit of that goal, recordings were made of the singers in the YADP on the premises of the Metropolitan Opera. These non-invasive measurements consisted of simultaneous audio and electroglottograph signals of a standardized protocol, comprised of familiar exercises (scales, arpeggios, sustained tones), as well as a voice range profile and portions of an individually selected aria. Each singer appeared for a 45-minute session in which the signals were taken on a two-channel DAT recorder for later analysis using the program VoceVista, developed by the authors at the Groningen Voice Research Lab in the Netherlands.
The opportunity to record signals of elite singers under controlled conditions is extraordinary, presenting the possibility of accumulating a database that can serve for other singing voices as a standard of comparison. The challenge of this voice documentation program is to find readily identifiable, objectively measurable points of comparison among singing voices that correspond to, and help explain, the qualitative differences apparent to subjective hearing. An intermediate goal to that end is to design a practical protocol of singing maneuvers that reveal these points under comparable circumstances for voices of widely varying types.
Results
The initial series of measurements was made on seven singers: three sopranos, two mezzo-sopranos, and two tenors. We make no attempt to generalize from such a small sample of these highly select singers, but this poster display presents some examples of the points of analysis that will be used in documenting the individual voices at this stage of their careers. Each example consists of an audio spectrogram of a short segment of the protocol, together with power spectra of the audio or waveform displays of audio and EGG signals, accurately aligned for simultaneity. In each case quantifiable, objective elements are identified that are considered characteristic of a given subject's voice. Audio recordings of the segments displayed are available for listening.