AMBIENT SOUND FOR CITY VR TOURS
BACKGROUND FOR A CITY
by Rich Studer
Our ears are the home of super-sophisticated mechanisms that let us maintain the standing position and capture an important part of the messages Mother Nature sends us constantly.
A brief biological note: the ear is the first sensitive organ completely formed in the fetus; the baby discovers the world and interacts with his mother, thanks to the vibrating stimuli he gets through his body and his ears. Listening is the ancestor of our communication tools. It is deeply imprinted in our consciousness.
Daily, we hear many sounds (maybe too many) but not all of them carry a significant message. We can decide when a sound becomes interesting and begin listening to it. We can isolate it from the magma and concentrate on its characteristics. That's the discriminating principle, which helps us to also determine the direction of a sound source.
This theoretical introduction may appear irrelevant, but in fact, it helps to understand the need of audio we prove in every kind of media. It's an unconscious adaptation to the real world. Cinema, TV and computer science tend to approach the reality with technical means.
VR also isn't new to the incorporation of multimedia content. QuickTime being one of the cleverest multi-format platforms, experimenting with multimedia has been a great pleasure for years.
The problem still worrying the content creators is the size of the media, especially if planned for use on the net, considering that a minute of a mono file weighs more or less 5 MB, non-compressed. Hi-bandwidth and MP3 improved the situation but still you can't put anything you want into your VR.
I invite you to consider the following example: VRWAY was building a network of multinode cities; the project was modular and destined to grow in size and in number of nodes. Our task was to demonstrate the use of audio into VRs for the people not accustomed to the media, and to give an example on how useful a soundtrack can be for the virtual traveler.
We started our brainstorming by analyzing this second didactic point. We imagined on each VR a speech, like a guided museum tour, we thought about hotspots with information about particular objects of the picture, and we planned a system of alarms and bleeps to give the directions to the user. In fact we gave to audio the role of message carrier.
These were good ideas but we had two impressive enemies: the short realization time and the impossibility for low-bandwidth users to jump into such a VR in less than 5 minutes! So we restarted, thinking more simply about audio's performance capabilities over that of written text. What could audio perform that written text couldn't? And we arrived at the principle of the
'rotating webcam' - a VR with the original sound in the place where the picture had been taken. This solution was of some merit as of the discrimination principle. It restored the reality in a VR, picture and sound. But we weren't enthusiasts yet and the realization time still would have been extremely long, as each VR would need to be post-scored.
This brought us to the definitive "postcard" model. A reasoned assembly of a city's own sounds, squeezed in a minute long file accompanying all the VRs belonging to that city. In this way, we preserved the discrimination factor, each city having its own soundtrack and differing from the others in many ways, the most noticeable being the spoken language. We defined a list of discriminating places and elements and went to the field to record our first city, Copenhagen.
In the example we collected three cities: Venice, Barcelona and Copenhagen
The experience taught us the following:
A VR with the soundtrack comes to life. It is much more precious and profound. More messages can be sent simultaneously through sounds.
It is possible to identify a VR with a soundtrack. If the nature of the sounds is real, the feelings come closer to the represented place. If it is synthetic, like music for example, the discrimination disappears and the attention will be focused more on the visual part. This second case may be useful when trying to collect different contents under one common identifier.
The soundcard model is a hybrid of the cumulative (establishing groups) and the discriminating principles. It's cumulative because it lays under all the VRs in looped mode, creating an aggregating ensemble and at the same time it's discriminating because sounds are proper to that situation. As an example, listen to the bells in Venice and in Copenhagen. Both are church bells but they sound different. Native inhabitants immediately recognize their own sounds.
Audio and video being heavyweight media, it is technically not evident to load many sounds onto a VR, especially when destined for Internet purposes. Programs like LiveStage easily allow the use of many sounds but they must be very short. A background sound with many "clicks" would be ideal.
A simple and light way to post-score remains MIDI. It uses QuickTime's internal generators and for basic purposes does a great job, intended mainly for music soundtracks.
Next issue I will discuss directional sound in a VR.
Rich Studer is a partner of a production studio -Digilab, owns a recording label ? Invisible Recordings ? and is a confirmed audiophile and electronic music fan.
Email: rich@invisiblerecordings.com
|  | |
| The purpose of this banner is to raise funds for a new VR community project VRMag will launch in a few months.
| |
