Audio-zoom
AudioZoom was founded in 2008 with the mission to develop a unique optoelectronic microphone, which will finally enable the mass-market usage of applications such as voice recognition, video conferencing, and hands free dialing/operation. The potential markets for these applications are enormous, and represent a market potential of several hundred $millions for AudioZoom’s products.
AudioZoom’s unique technology creates a “Virtual Cube” in space, sensing sound only from within the cube. This way, the only sound sampled is from the relevant speaker, thus overcoming the root cause for ambient noise.
The result is a near-zero noise output and precise speaker separation, which are the key elements missing today to enable mass-usage of voice applications.
Voice is a fundamental way of communication, and has the potential to be the most pervasive human-machine interface. For example, every computer user would greatly benefit from the ability to dictate and run applications by voice commands, while mobile users would be able to send SMS and email. Evident to this potential are the many attempts made over the past decades to provide voice recognition systems, which would harness voice in order to increase productivity.
Premature deployment of speech recognition systems in the 80’s and 90’s has limited market penetration, but there is evidence that market perceptions regarding speech recognition products, functionality, and ease of use are beginning to change. In 2006 the overall market for ASR technology topped $1 bln (an increase of 100% in just two years), and sales soared to $4 bln in 2009.
One of the main problems that still limit the usage of such systems, and that has not been overcome to date, is the poor quality of sampled voice due to surrounding noise including voices from additional speakers and ambient noise. In fact, most voice applications today cannot be used in open space, due to the issue of surrounding noise.
Various attempts have been made to overcome surrounding noise by use of filtering and signal processing technologies, in order to isolate the voice from the target speaker. These attempts have an inherent disadvantage (using the same acoustic information for filtering and voice output), and thus have resulted in systems that require quiet surroundings as well as use of very high-end microphones.
-
Tal Bakish, Founder & CEO: 11 yrs experience in VLSI engineering with Cisco Systems, IBM, Modu-mobile, Intel, Connect-One, Transwitch and Compass-EOS, and Founder/Investor in several startups. B.Sc in Physics
-
Yechiel Kurtz, Active Director/Investor: 22 yrs of operational and investment experience as an executive with Apax Partners and MRV Communications. Involved as an Angel investor in several startups, from inception to exit. MBA and a BScEE.
-
Gabi Horovitz, VP R&D: 24 yrs of R&D practice, with unique multidisciplinary experience in optics, electronics and signal processing algorithms. Proven success in developing a product from concept to commercial product. MsCEE and BScEE.
-
Dr. Yekutiel Avargel: Signal processing developer: Research Engineer with RAFAEL Research Laboratories. B.Sc., M.Sc., and Ph.D. in EE
-
Prof. Israel Cohen, Signal Processing Advisor: Considered to be one of the leading authorities worldwide on signal processing for speech enhancement
AudioZoom’s unique solution overcomes these issues by combining an acoustic microphone together with an optical sensor that samples face vibrations. This unique (patent pending) combination creates a “Virtual Cube” around the target speaker, with the following characteristics:
-
Near-zero noise, including separation of the active speaker from other speakers
-
Fully directional, containing voice only from the relevant speaker
-
Precise Voice Activity Detection (VAD)
AudioZoom has completed Proof Of Concept prototype which has verified technology and the commercialization aspects.
The company has finaized first working prototype and now approching leading potential customers. The compnay is seeking investments for commercialization of the product and for bussiness development activities.
The company’s go-to-market will focus on mass market penetration via strategic relationships with large OEMs for integrated/private label products, such miniaturized microphones for OEM integration (Netbooks, car kits, hearing aids, etc.) and aftermarket products (PC USB/Bluetooth microphones, etc.).