Tel Aviv — Israeli start-up Hi Auto will unveil the world’s first audio-visual commer­cial solu­tion for back­ground noise-free spea­ker recor­ding and speech enhance­ment. The company has comple­ted a seed phase with a $4.5 million invest­ment led by Israeli auto importer Delek Motors and Hi Auto’s acting chair­man Zohar Zisa­pel (pictu­red).

Other inves­tors included Allied Holdings (holding company of auto importer Cham­pion Motors), Singa­­pore-based Gold­bell Group, a leader in indus­trial vehicle sales and leasing, and Plug & Play, which works with auto­mo­tive part­ners to innovate.

Hi Auto plans to use the seed stage funds to complete deve­lo­p­ment of its first product, launch distri­bu­tion and expand its team.

Speech reco­gni­tion plat­forms are curr­ently strugg­ling with the problem of poor reco­gni­tion rates in noisy envi­ron­ments, such as when there are child­ren in the back seat of a car, an ambu­lance drives by, or it is raining heavily. Hi Auto’s audio­vi­sual solu­tion elimi­na­tes all noise and ensu­res that the speech reco­gni­tion plat­form works relia­bly in all noise conditions.

The novel solu­tion can be used in a wide range of use cases in various fields where speech reco­gni­tion and clearly intel­li­gi­ble conver­sa­ti­ons are requi­red, such as in the auto­mo­tive indus­try, cont­act centers, video confe­ren­cing, or robots serving custo­mers in the hospi­ta­lity indus­try. Curr­ently, Hi Auto is focu­sed on the auto­mo­tive sector and envi­si­ons the new tech­no­logy as a solu­tion to the problem of voice-based vehicle control.

Hi Auto is based in Tel Aviv and was foun­ded in Febru­ary 2019 by CEO Roy Baha­rav, CTO Eyal Shapira and Zohar Zisa­pel, who serves as the company’s acting chair­man and inves­tor. Baha­rav recently retur­ned to Israel after living on the West Coast of the United States for 12 years. There he held various product manage­ment posi­ti­ons at Google’s Moun­tain View head­quar­ters in Cali­for­nia, as well as CEO of SeamBI, a company he foun­ded in 2006. In the past, he held senior posi­ti­ons in an elite intel­li­gence unit of the Israel Defense Forces (IDF), such as comman­der of a divi­sion of 40 engi­neers and tech­ni­cal experts.

Eyal Shapira also comman­ded a detach­ment of some 40 engi­neers and specia­lists in an elite IDF tech­no­logy unit, in which he served and was awarded the Israel Defense Prize. After his mili­tary service, Shapira was a foun­der and admi­nis­tra­tor of start-up compa­nies and a consul­tant for compa­nies such as Intel and Broad­com. Zohar Zisa­pel, a major inves­tor in the auto­mo­tive sector, is one of Israel’s best-known high-tech entre­pre­neurs and mana­gers. Zisa­pel helped found and is chair­man of some of Israel’s best-known auto­mo­tive start-ups, inclu­ding Hailo, Argus and Innoviz.

Hi Auto will unveil a proto­type of the world’s first commer­cial driver voice reco­gni­tion solu­tion, which sepa­ra­tes the driver’s voice from all other voices in the vehicle, as well as noises inside and outside the vehicle, or hides all back­ground noise, at CES 2020 in Las Vegas from Jan. 7–10. Thus, for the first time, such a solu­tion is coming to the market that combi­nes a micro­phone placed in front of the inten­ded spea­ker with a camera that tracks his or her lips. The system is based on a deep-lear­­ning soft­ware program instal­led on the device that elimi­na­tes noise, which cannot be achie­ved by using audio tech­ni­ques alone. The company has alre­ady estab­lished cont­acts with leading auto­mo­tive compa­nies, which are expec­ted to test its tech­no­logy in early 2020.

“Whether for use in purcha­sing train tickets, navi­ga­tion control or lane change commands, voice reco­gni­tion is rapidly beco­ming the most popu­lar device control method almost ever­y­where in the world,” said Roy Baha­rav, CEO and co-foun­­der of Hi Auto. “Howe­ver, when the devices are opera­ted in envi­ron­ments with multi­ple spea­k­ers or in noisy envi­ron­ments, their relia­bi­lity decrea­ses drama­ti­cally. Our audio­vi­sual solu­tion is able to focus on the spea­ker and remove all back­ground noise, making it sound like they are in a recor­ding studio. Our solu­tion will make the expe­ri­ence around speech reco­gni­tion in the car, as well as in other envi­ron­ments, more satis­fy­ing for consu­mers and enable the intro­duc­tion of more complex and sensi­tive features by origi­nal equip­ment manufacturers.”