The AudioCamera’s spherical microphone architecture provides an omni-directional acoustic image of the environment.
The microphone signals are conditioned using individualized pre-amp circuits and then converted to digital by a dedicated analog to digital converter. The digitized acoustic data are sampled at 44.1 kHz per channel and collected by an internal high performance Field Programmable Gate Array (FPGA) based processor where they are managed, formatted and converted into a single USB 3.0 data stream. FINDr software, operating on a standard PC with an internal NVIDIA GPU, accepts the USB data stream and performs the user selected functions in real-time.
The auditory scene can be decomposed into its spherical harmonic components up to order 7. This allows for increased spatial isolation of acoustic sources in the environment. Additionally, the scene can be decomposed into filtered plane-waves, and can be beamformed into multiple real-time beams. A resolution of up to two degrees is possible for higher frequency beams.
The video camera array generates an associated omni-directional panoramic visual scene that is synchronized with the live audio stream.
Five video cameras are integrated into the VisiSonics 5/64 Audio/Visual Camera to facilitate the panoramic scene capture. Each of the color video cameras has a resolution of 1328×1048 pixels. The captured video is synchronized with the audio data via an internal FPGA processor. The multi-camera image streams are collected via an internal USB 3.0 interface and sent to a laptop over a single USB 3.0 cable.
Included with the all Audio Cameras is the Frequency Identification and Noise Detector (FINDr) software.
With FINDr, you have the ability to visualize and analyze your recorded data. The FINDr software gives you instant access to an integrated audio/video panorama, and breaks down the data for you in helpful graphs for both time and frequency domain analysis.
See our Software page for more information.