VocalZoom and iFLYTEK Partner to Give China’s Most Popular Speech Recognition Platform Breakthrough Performance in Noisy Environments

News Highlights:

  • Early tests show VocalZoom’s optical sensor dramatically improves automatic speech recognition (ASR) performance of iFLYTEK’s voice-control platform by an average of more than 50 percent.
  • Greatest performance enhancement is achieved in environments with the highest noise levels, such as driving an automobile with the windows rolled down.
  • Companies are exploring combination of VocalZoom sensor technology with iFLYTEK’s Voice Cloud, the leading intelligent speech platform in China for mobile cloud and embedded applications.

Yokneam Illit, Israel and Hefei, Anhui, China – January 25, 2016VocalZoom, a leading supplier of Human-to-Machine Communication (HMC) optical sensors that enable a more natural, personalized and secure voice-control experience, today announced it has signed an agreement with iFLYTEK to combine and test the performance of VocalZoom’s optical HMC sensor with iFLYTEK’s Voice Cloud intelligent speech technology platform, China’s most widely deployed solution with hundreds of millions of users. Initial results show that the ASR performance of iFLYTEK’s platform can be improved an average of 50 percent, and even more in noisy environments, by adding VocalZoom sensors to user headsets and automotive infotainment solutions.

“iFLYTEK is already the best-performing voice-control solution by a number of global standards across all popular benchmark test domains, but the biggest challenge is noisy environments and we believe that VocalZoom technology gives us the foundation for breakthrough improvements,” said Haikun Wang, senior researcher at iFLYTEK. “We have tested the VocalZoom sensor in scenarios including the inside of a moving automobile with the windows rolled down and high wind noise, and the improvement is quite substantial.”

The key for human-to-machine interactions using virtual assistants and automotive voice control is whether the machine did what it was told, quickly and accurately enough to satisfy the user. Even with the latest noise reduction algorithms, today’s acoustic microphones can’t achieve adequate voice isolation for this level of control, especially in noisy environments. The VocalZoom multifunction HMC sensor overcomes this problem by gathering additional data generated during speech as facial skin vibrates around the mouth, lips, cheeks and throat. By integrating the VocalZoom optical HMC sensor into a voice-control solution and focusing it on these areas, facial vibrations can be acquired, measured and converted to an isolated, near-perfect reference signal with which the system can operate – regardless of noise levels.

“We are excited to be working with iFLYTEK to significantly improve the performance of one of the most widely deployed ASR platforms,” said Rammy Bahalul, vice president of sales and business development for VocalZoom. “VocalZoom and iFLYTEK intend to complete our joint testing this month as we explore opportunities to combine our two technologies into a revolutionary new solution for voice control and authentication. We share the vision of a significantly better voice-control experience using smaller, sleeker headsets and other solutions across a wide variety of applications.”

VocalZoom and iFLYTEK have tested the performance of a solution combining both companies’ technologies inside a moving automobile across a number of scenarios with varying levels and combinations of music, car noise and other interference. ASR performance is improved to an almost perfect score in environments where high noise previously made adequate voice-control performance impossible.

About VocalZoom

VocalZoom supplies Human-to-Machine Communication (HMC) sensors for delivering a, natural, personalized and secure voice-controlled user experience in today’s increasingly mobile and interconnected world. The sensors enable accurate and reliable voice control and biometrics authentication in any environment, regardless of noise. Applications including mobile secure payments, headsets and wearables, mobile phones, access control, smart home solutions, and hands-free automotive voice control. For more information, visit www.VocalZoom.com and follow the company on LinkedIn.

 

About iFLYTEK

Established in 1999, iFLYTEK is a national level high-tech enterprise dedicated to the research and development of intelligent speech and language technologies, artificial intelligence, hardware & software applications, and provision of professional services for governments, education sector, financial organizations and other fields. iFLYTEK was listed in the Shenzhen Stock Exchange in 2008 (stock code: 002230). iFLYTEK’s intelligent speech and artificial intelligence technologies such as speech synthesis, speech recognition, speech evaluation, and natural language processing, represents the top level in the world. iFLYTEK has occupied more than 70% of Chinese speech industry market share. Meanwhile, iFLYTEK provides speech core technology for more than 2,000 companies in the whole industry and has launched the world’s first “iFLYTEK voice cloud” platform, which provides intelligent speech interaction capability for mobile internet industry(currently there are more than 80,000 project partners and more than 700 million end-users, promoting the application of intelligent speech and artificial intelligence technology into education, mobile phones, automotive, appliance and other industries, serving millions of households. For more information, visit iflytek.com.