The Chinese startup iFlytek is hoping to introduce its voice recognition platform at a national scale within the next two to three years. The company leverages AI and big data to identify users based on unique vocal traits, and is now claiming to have made a research breakthrough that will enable a more widespread deployment in China.
According to iFlytek, the platform will improve call center security in finance and other industries. The technology allows companies to monitor phone calls in real time, using voiceprint recognition for ongoing biometric authentication. iFlytek is currently trialing a version of the platform that will automatically terminate a phone call if it suspects a scammer.
The technology will also help law enforcement investigate crimes that have already occurred. Specifically, voiceprint recognition will theoretically allow officers to identify and catch phone scammers based on recordings of their activity.
“Because recordings are important evidence when it comes to phone scams, demand for voice recognition is growing,” said Fu Zhonghua, the deputy head of the iFlytek research center in Xi’an.
iFlytek is hoping that domestic demand for voice recognition will boost its prospects within the country and offset U.S. sanctions that were introduced in response to human rights violations of Muslim minorities in the Xinjiang region. iFlytek is one of 28 Chinese entities that was placed on a U.S. blacklist for participating in the surveillance program in the area.
However, the company may not mind too much if it is able to capture the massive Chinese market, and iFlytek is well-positioned to do so after receiving approximately $39.4 million in government subsidies in 2018 alone. The company has previously provided speech recognition software for the digital signal processing company CEVA, which is based in California.
Source: Nikkei Asian Review
January 3, 2020 – by Eric Weiss