University of Pennsylvania, 3401-C Walnut Street, Suite 300, C Wing, Philadelphia, PA, 19104, United States
Jianjing Kuang
Popular version of 4aMU6 – Ultrasound tongue imaging of vowel spaces across pitches in singing
Presented at the 186 ASA Meeting
Read the abstract at https://doi.org/10.1121/10.0027410
–The research described in this Acoustics Lay Language Paper may not have yet been peer reviewed–
Singing isn’t just for the stage – everyone enjoys finding their voices in songs, regardless of whether they are performing in an auditorium or merely humming in the shower. Singing well is more than just hitting the right notes, it’s also about using your voice as an instrument effectively. One technique that professional opera singers master is to change how they pronounce their vowels based on the pitch they are singing. But why do singers change their vowels? Is it only to sound more beautiful, or is it necessary to hit these higher notes?
We explore this question by studying what non-professional singers do – if it is necessary to change the vowels to reach higher notes, then non-professional singers will also do the same at higher notes. The participants were asked to sing various English vowels across their pitch range, much like a vocal warm-up exercise. These vowels included [i] (like “beat”), [ɛ] (like “bet”), [æ] (like “bat”), [ɑ] (like “bot”), and [u] (like “boot”). Since vowels are made by different tongue gestures, we used ultrasound imaging to capture images of the participants’ tongue positions as they sang. This allowed us to see how the tongue moved across different pitches and vowels.
We found that participants who managed to sing more pitches did indeed adjust their tongue shapes when reaching high notes. Even when isolating the participants who said they have never sung in choir or acapella group contexts, the trend still stands. Those who are able to sing at higher pitches try to adjust their vowels at higher pitches. In contrast, participants who cannot sing a wide pitch range generally do not change their vowels based on pitch.
We then compared this to pilot data from an operatic soprano, who showed gradual adjustments in tongue positions across her whole pitch range, effectively neutralising the differences between vowels at her highest pitches. In other words, all the vowels at her highest pitches sounded very similar to each other.
Overall, these findings suggest that maybe changing our mouth shape and tongue position is necessary when singing high pitches. The way singers modify their vowels could be an essential part of achieving a well-balanced, efficient voice, especially for hitting high notes. By better understanding how vowels and pitch interact with each other, this research opens the door to further studies on how singers use their vocal instruments and what are the keys to effective voice production. Together, this research offers insights into not only our appreciation for the art of singing, but also into the complex mechanisms of human vocal production.
Video 1: Example of sung vowels at relatively lower pitches.
Video 2: Example of sung vowels at relatively higher pitches.
Michael Oelze – oelze@illinois.edu
X (Twitter): @Oelze_Url
University of Illinois at Urbana-Champaign
Urbana, IL 61801
United States
Zhengchang Kou
University of Illinois at Urbana-Champaign
Urbana, IL 61801
United States
Popular version of 1pBAb7 – Contrast-Free Microvessel Imaging Using Null Subtraction Imaging Combined with Harmonic Imaging
Presented at the 186th ASA Meeting
Read the abstract at https://doi.org/10.1121/10.0026780
–The research described in this Acoustics Lay Language Paper may not have yet been peer reviewed–
Bubbles? We don’t need no stinking bubbles!
In recent years, super resolution imaging techniques for imaging the microvasculature have been developed and demonstrated for applications such as functional ultrasound imaging in mice and assessing Alzheimer’s disease in pre-clinical models. While these novel super resolution techniques have produced images with incredible detail and vessel contrast, one drawback to the approach is the need to inject contrast agents, which consist of small gas-filled microbubbles. This reduces the clinical application and adoption of these techniques. Furthermore, the time required to construct these images can take hours because it involves localizing and tracking individual microbubbles as they progress through the vasculature.
FIGURE 1: Video of 3D rendering of rat brain using traditional approaches at a fundamental frequency (top left) and at twice the fundamental frequency (bottom left) compared to using our novel approach (NSI) at a fundamental frequency (top right) and at twice the fundamental frequency (bottom right) (please note this will be a playable video online)
In our novel approach to microvessel imaging, we don’t need no stinking microbubbles! Instead, we utilize a novel nonlinear beamforming approach that allows fast reconstructions with much better spatial resolution. This allows us to approach super resolution without the need to inject microbubbles into the body. Along with the beamforming approach we also use a pulse inversion scheme, where we transmit with one frequency and receive with twice the transmit frequency. This allows a doubling of the spatial resolution over receiving with the same transmit frequency. However, the use of pulse inversion scheme can introduce unwanted clutter into the image. With our novel beamforming approach, clutter is greatly reduced or eliminated from the images.
FIGURE 2 Single image frames comparing traditional power Doppler methods (left) with our novel approach (right).
We demonstrated our new technology in a rat brain (both 2D and a 3D rendering) and rabbit kidney and compared our images to traditional beamforming approaches without the use of contrast agents. The video shows a 3D rendering of the microvasculature of a rat brain and the corresponding figure shows a particular frame of the 3D rendering. We showed that our approach eliminates the clutter produced by the pulse inversion scheme, increases the contrast of microvessel images, results in more observable vessels, and produces a much finer spatial resolution better than one fourth of a wavelength. The time to reconstruct the images using our novel technique was a fraction of the time needed for current super resolution techniques that rely on localizing and tracking microbubbles in the vasculature. Therefore, our novel approach could provide microbubble-free technology to produce high-resolution power Doppler images of the microvasculature with the potential for clinical applications.
FIGURE 2 Single image frames comparing traditional power Doppler methods (left) with our novel approach (right).