The Transgendered Voice: Beyond the Pitch Change
James Dembowski - james.dembowski@ttuhsc.edu
Texas Tech University Health Sciences Center
3601 4th Street
Lubbock, TX, 79430-6073
Popular version of paper 1aSC6
Presented Monday morning, October 26, 2009
158th ASA Meeting, San Antonio, TX
Everyone knows that women tend to have higher pitched voices than men. But is it primarily vocal pitch that makes a woman sound like a woman and a man sound like a man? When men try to imitate women, they raise the pitch of their voice. But when you hear a man do this, do you ever really think youre hearing a woman? Or rather, do you think, That sounds like a guy trying to talk like a woman? Think about it for a moment, and youll realize that men who raise their voices to imitate women dont really sound much like women at all. And that should tell you that while it is true that women tend to have higher voices than men, it takes much more than a high voice to make a woman sound womanly. In fact, a woman may sound very much like a woman even without a high voice. Sometimes we particularly value women with low pitched voices, and consider their voices exceptionally alluring. Think of Lauren Bacall or Greta Garbo, or any number of female broadcasters on National Public Radio.
This raises a thorny set of issues for speech pathologists or voice coaches who find themselves treating transgendered clients who have transitioned from male to female. First of all, producing a high pitched voice in a person who passed puberty as a male is not easy. Hormone therapies used by transgendered women produce a number of secondary female characteristics, but they do not change the sound of the voice. In order to do so, a transgendered woman must habituate a set of vocal techniques to raise her voice; these techniques may not be comfortable for her, and the process of habituation can be an arduous one. Second, even if a woman succeeds in raising the pitch of her voice, that does not guarantee that she will sound like a woman. Finally, what if a woman does not want a high pitched voice? A mature, successful, professional woman in a position of authority may find that a stereotypically high pitched female voice is not desirable. This is just the case in the study presented here.
This case study shows preliminary results of acoustic analysis of the speech and voice of one male to female transgendered woman. Following hormone therapy and gender reassignment surgery, she came to the clinic interested in developing a more feminine voice. Ms. J was highly articulate, had a successful academic career, was an enthusiastic teacher, and was responsible for substantial administrative duties. She was an authoritative woman who had no desire to express herself in a high pitched voice, which would have been inappropriate for a woman of her position. The challenge was to achieve a feminine voice perhaps a better term would be a womanly voice without sacrificing authority and without focusing on pitch.
We evaluated Ms. Js speech to find characteristics other than pitch which might be altered to give a perception of femininity. One striking characteristic of her speech was that she spoke at a high speech rate, sometimes in excess of 7 syllables per second. A more conventional speech rate is 3 4 syllables per second, and there is some research to suggest that women tend to speak more slowly than men. Thus, one goal was simply to slow Ms. Js speech rate. Not only would this give her a speech rate closer to that of a typical woman, but it would give her more time to focus on the sound of her voice as she spoke. Ms. J also tended to produce a great deal of the voice quality known as glottal fry. This is a low-frequency, growly sort of voice that almost everyone produces at some times (at the ends of long sentences, for example, when were running out of air), but which Ms. J produced excessively. By reducing glottal fry, we hoped to accomplish several things. First, we would eliminate some of the lowest frequencies in her speech. Note that this is not the same as raising pitch. Its the difference between giving someone a high pitched voice (which we did not try to do), and selectively removing some of the lowest pitched components of a voice (which we did try to do). By reducing glottal fry, we also hoped to facilitate production of a gentler, breathier quality of voice. Finally, we worked on changing Ms. Js patterns of marking stress. Men tend to mark stress more by changing the intensity (or loudness) of their words, while women more often mark stress by changing the pitch (the melody) of their words.
After six months of therapy, we were able to move Ms. Js speech patterns in the desired direction in some respects, but not in others. In reading samples, Ms. J showed a reduction in average speech rate from 5.8 syllables per second to 4.4 syllables per second. Ms. J also showed an increase in her average fundamental frequency (F0), the rate of vocal fold vibration, which correlates with vocal pitch. For adult men, this tends to be about 120 times per second, and for adult woman about 200 times per second. From early to late reading samples, Ms. J raised her average F0 from 127 cycles per second to 146 cycles per second. Thus, her average vocal pitch came close to the upper part of the male range. Remember, however, that we were not specifically trying to raise pitch. Why then this increase? We believe that it was because we had some success in modifying her use of glottal fry. That is, she eliminated some of her lowest, most growly pitches, raised the floor of her frequency range, and reduced the number of occurrences of glottal fry. We were less successful in altering the pitch variability of her speech (introducing a more melodic contour), and in introducing a breathier overall quality.We are still working toward these goals.
The take-home message of this report is that producing a feminine voice is not primarily a matter of raising pitch. Rather, a host of acoustic variables go into the production and perception of a womans voice. Raising pitch to increase a perception of femininity may not only be difficult, it may not even be desirable. In these cases, therapists need to produce changes of speech rate, vocal quality, and prosody. The gender-related acoustics of these features have not been studied as extensively as vocal pitch, and the work of speech pathologists and voice coaches could be facilitated by further quantitative acoustic research into features other than pitch which produce a womanly voice.