ANU researchers have developed a new type of microphone array capable of recording 3D spatial sound, using a planar geometry.
Capture and reproduction of 3D audio is becoming increasingly important for many applications including augmented reality, virtual reality, media, digital entertainment, human-machine communications/interfaces, smart homes, hearables, wearables, medical aids, teleconferencing and active noise control within confined spaces. Currently, binaural, hemispherical or spherical microphone arrays are used to capture 3D sound. Binaural arrays consist of dual microphones that are shaped and positioned like human ears. Hemispherical arrays typically involve 2-14 microphone capsules arranged in a tetrahedral pattern, whilst spherical arrays incorporate numerous pressure microphones which are mounted on the surface of a rigid spherical-shaped baffle. The spherical array geometry has several advantages over the other geometries, e.g. the polar pattern can be directed to any direction in the 3D space without changing the shape of the pattern, plus the spherical array allows full 3D control of the polar pattern and spatial-filtering in 3D soundfields. The downside, however, is that this geometry is bulky in nature, which makes this type of array non-portable and inconvenient for practical use. Spherical arrays are widely used for research and experimental purposes, but the size greatly limits the arrays' potential in commercial applications. ANU researchers are seeking to address this limitation by developing a microphone array with reduced dimensionality and size that will still offer the same functionality and benefits of the spherical microphone array.
ANU researchers have developed a novel, compact microphone array arrangement and custom digital signal processor algorithms for enhanced 3D sound capture and delivery. Through the use of directional microphones, the researchers have been able to significantly reduce the size of the array, as compared to conventional spherical 3D arrays, which have similar functionality.
Currently, the team is exploring the use of the array in desktop microphone (Figure 3B) for hearing aid users. The intended end-product is a compact, wearable (or desktop) array that captures the full spatial sound information around the hearing aid user, then a dedicated binaural rendering algorithm processes the captured ambisonic signals to generate two channels of audio output, which are sent to each of the user's hearing aids.
- Compact for easier integration into consumer electronics
- High quality 3D audio, inc. removes unwanted background noise and automatically focuses on desired sound/speech source
- Adaptable, with the ability to changes quality of components
- Customisable to a ranges of devices and applications
The planar microphone array technology is suitable for a range of devices and applications and could benefit a variety of company types - as such it may present multiple potential licensing opportunities. ANU is interested in identifying industry partners that have a need for this technology, in order to create a number of functional prototypes. ANU is well placed to work with partners to optimize an array for their specific application.