The tests prior to production were done with a Matlab code called VTCalcs. This program allowed me to adjust 30 different sliders, representative of dividing the vocal tract into 30 sections, to recreate the area function of each vowel. Using this information Matlab synthesized the sound, estimated the formant frequencies, and provided an estimate of the tract contour.
For a closer look click on the desired image!
![Beet](https://muse.union.edu/griffiths-capstone/files/2015/03/Beet.jpg)
![Bit](https://muse.union.edu/griffiths-capstone/files/2015/03/Bit.jpg)
![Bet](https://muse.union.edu/griffiths-capstone/files/2015/03/Bet.jpg)
![Foot](https://muse.union.edu/griffiths-capstone/files/2015/03/Foot.jpg)
![Boot](https://muse.union.edu/griffiths-capstone/files/2015/03/Boot.jpg)
The results were then compiled and compared to the standard values of each formant frequency.