NSF Award
1430908
The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project extends to wherever headphones are used to listen to sound. The first commercial markets targeted by the technology being developed are gaming, virtual reality and consumer entertainment. These are areas in which the US has strong market and technology leads. By the end of the project, the company intends to have developed software that can be licensed to major players in these areas, reaching tens of millions of users. Beyond these markets, there are many applications for spatial audio in human-computer interfaces to deliver spatial information along with the intended semantic message. Several niche markets also exist, such as data presentation/exploration via sonification and specifically designed auditory interfaces for vision-impaired users. Overall it is expected that the proposed R&D work will advance the state of the art in science and technology and will be of substantial value to society as a whole due to high usability, fidelity, naturalness, and portability of the developed spatial audio solutions.<br/><br/>This Small Business Innovation Research (SBIR) Phase II project seeks to develop highly realistic and computationally efficient software for synthesis of personalized spatial audio, for applications that include virtual reality, gaming, and prostheses for the blind. Current algorithms are either perceptually unsatisfactory or have very high computational load. Approximate modeling of sound propagation, reverberation, and diffusion, along with tradeoffs between complexity and quality, will be explored using perceptual distortion metrics combined with the skills of professional listeners. The software will be optimized for use on mobile, console and embedded platforms. Additionally, a method to personalize the software to individual listeners, which was previously developed and tested in laboratory conditions, will be further refined and ruggedized for use in realistic environments. The expected outcomes will include a high-quality and efficient audio rendering library, a portable personalization apparatus, and several demonstrations that highlight the capabilities of the technology.
