A sample silicone vocal fold on a human hand.
Embedded Files
Pitch Control of Multi-Layered Self Oscillating Vocal Folds Using Shape Memory Alloy
Pitch Control of Multi-Layered Self Oscillating Vocal Folds Using Shape Memory Alloy
Fall 2022
PURPOSE
To replicate human vocal folds and cords using silicone rubber and apply tension using shape memory alloys (SMAs).
Terminology & Definitions
Terminology & Definitions
SMA - Shape Memory Alloy. When heated or when electricity is sent through, an SMA will shrink. When cooled, it will expand to its original shape.
Op-Amp - Operational Amplifier. An integrated circuit that amplifies the difference in voltage between two inputs.
CONTEXT & MOTIVATION
Millions of Americans experience vocal disorders, such as aphonia. Mechanically modeling vocal cords offers a unique solution to artificially producing sound that could bring full functionality of the voice back to those that suffer from these diseases.
The human vocal system can be split up into four main components:
Actuation - The breath. The energy of the voice.
Vibration - The pair of vocal folds oscillate due to air pressure coming from the lungs.
Resonance - Occurs in the vocal tract. Amplifies the sound of one's voice.
Articulation - The tongue and lips shape sound into syllables.
For the purposes of this study, we focused mainly on recreating the first three phases, as articulation adds additional layers of complexity.
Previous research was done around the concept of mimicking the vocal system. One such paper successfully controlled pitch to an accuracy of approximately +/- 3 Hz using servo motors. Our project not only involved recreating the vocal system, but also implemented SMAs in the tension of vocal cords due to potential quieter actuation, improved precision, and their use in biomechanical applications.
EXPERIMENT OVERVIEW
Two artificial vocal folds were placed over an air tube, where pressurized air was sent through. SMAs were used to pull elastic threads that passed through the vocal folds, causing them to tense and thus altering the pitch of the sound produced. The elastic threads also represented the elastic fibers in the vocal cords. The pitched sound then traveled through a resonance tube and finally into a microphone. Sound data was then collected and analyzed to determine the accuracy of the changes in pitch.
The figure to the right depicts the different parts of the mount. The vocal folds, which can be seen in pink, were attached to a pair of acrylic mounts, which are colored in green. The mounts were connected by a long bolt, but held apart at a small distance using cell foam. The cell foam, in blue, allowed for tension and compression between the plates while also preventing air from escaping through anywhere other than between the vocal folds. The air that did blow past the vocal folds produced a sound that traveled through the resonance tube, which is highlighted in purple.
This mount assembly was placed on an 80/20 structure that allowed for the integration of the other subsystems. The green line at the bottom of the structure represents the air tube, where air flowed from an air compressor at a controlled pressure. Tension on the elastic threads within the vocal folds was applied through SMAs, which controlled using an Arduino. The microphone above the resonance tube collected sound data through LabVIEW.
RESULTS
Ultimately, we were successful in changing the pitch of the sound created by our vocal fold replicas.
However, we discovered that the SMAs currently available for research purposes were not sensitive enough to be used for this purpose. In order to yield better results, the experiment should be redone in the future, when different alloys with less heat sensitivity are developed for use as SMAs.
Page updated
Google Sites
Report abuse