Kristal Cazella's Portfolio

Research paper: Virtual Sound Environment - 1532 words

Bring primarily interested in the creation of virtual visual environments, especially three-dimensional, I wanted to explore an application to audible space. This project initially started as an individual effort (see next tab, Creative: Virtual Room) and was going to implement 2-speaker (headphone) stereo. Toney Vecchio was researching spatial sound at the same time, and so we decided to collaborate by combining our efforts. Parts of my original patch were used and he introduced the sound package Ambisonics into the mix. Both of us would tweak the compilation together to better suit our needs. Features I had planned for my own patch were implemented here instead: most notably the controller interface for multiple sounds.

speakers.bmp
levelmeter.jpg

Setup
Instead of my originally intended 2-speaker stereo, Toney had done research into applications of 5.1 stereo, which would be handled by Ambisonics' ample channels. Within the senior sound lab five speakers were set up around a central point as pictured. Each number corresponds to the channel routed.
A sound level meter similar to the one pictured was used to balance the speakers. The process was as follows: The meter was set at a central point. Pink noise was generated and routed to each one of the speakers individually in turn. Each speaker was adjusted to a central volume level, resulting in a smooth balanced sound based in the spot where the meter was placed.

Development
On the outside, there is very little complexity to comprehend. The gray box represents your virtual room, with the gray dot representing yourself, your position (controlled by dragging or directly manipulating coordinates) (A gaming-controller was going to be implemented to manipulate position, but development on the internals became more of a priority as time grew short). Orientation is dictated on each side of the room, and it is expected that you as a listener maintain this orientation (no head turning). Each of the four colored circles represents a sound source. These can independently be manipulated in multiple ways using the available interface: position, autopan (sets the sound orbiting, regardless of position), and volume. Azimuth is visible and is an indicator of the angle relationship between the attached sound and yourself.

presentscreen

Essentially, once the patch was worked out for a single sound source (marked my purple within the patch), a bit of tweaking was needed to add a second, and then that second could be used as a repeatable template for any number of sources beyond.
Click here for a closer view!

Challenges
There were many challenges, problems, and bumps along the way to the current point of development. An expected obstacle was understanding how to use Ambisonics to accomplish our goals. Another annoyance was the frequent necessity for geometry- and trigonometry-related equations in order to translate data into usable numbers. An unexpected and rather frustrating barrier arose midway through development when an outside student tampered with the physical setup of the project. Progress became difficult to evaluate and the functionality of the equipment was called into question. Thankfully with the help of the sound level meter, balance was restored.

Ambisonics
Although I like to think I was on the right track, Ambisonic does what I was struggling to accomplish, and more! A lot better, too. It saved me a lot of work, though perhaps if I had had the time I would like to complete my own methods in order to compare. In the aforementioned screenshots, the Ambisonics components are marked by the darker violet.

Conclusion
Though Ambisonics was at first intimidating, after getting to know some of its capabilities and advantages I find myself motivated toward deeper and more complex applications, building off of the current patch. If a lot more time was available, I would have liked to continue to develop and expand the current version, adding features and pursuing the idea of a virtual room. Overall though, I am satisfied and pleased with the progress made. This patch could be applied in the future to other projects, perhaps, by myself or others.

Contact me: kcazella@stetson.edu

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License