Will voiceprint recognition become the new password after facial recognition?
By Goraud Mazanec 2019-07-17 448 0
There have also been recent reports that Siri will also add voiceprint recognition, and when you say "Hey Siri," Siri can know if the owner is calling it. This may not be just a rumor, after all, Apple posted on its official technology blog last year, disclosing the details of the voiceprint recognition technology for Siri.
The so-called voiceprint recognition (Voiceprint Recognition), is simply to identify the person who speaks by voice. Because of the differences in vocal organs, language habits, pronunciation size, frequency, etc., everyone's voiceprint map is unique, just like fingerprints, it is a biometric recognition technology that can be used for password authentication.
So after we experience from fingerprint recognition to facial recognition, will voice become the new mainstream password authentication mode?
Although voiceprint recognition is not yet popular in consumer products such as smartphones, it has been widely used in areas such as bank transactions and national security.
According to foreign media This is Money, Barclays, HSBC, Halifax and other British banks currently support voiceprint recognition, it is reported that more than 3 million bank customers in the UK use voiceprint recognition system to log in to their bank accounts.
You may not expect that the people most in favor of this new technology are the elderly. Some banks say there has been a surge in registrations in the elderly because voiceprint recognition eliminates the need to remember passwords and reduces the likelihood of fraud.
Michael Wallis, a retired music teacher in Hampshire, England, has been robbed of account information by a swindler and transferred from his account. The bank advised him to use voiceprint recognition technology. Michael has now used voiceprint recognition to make more than 100 transactions and check balances. He said it made him more comfortable.
The banks are also said to use voiceprint recognition technology to identify whether the sound source is a recording, in order to prevent swindlers from deceiving the system through recording.
However, there are still some problems with the technique, such as the Michael feedback system sometimes does not recognize his voice in the event of a sore throat or cold.
And like facial recognition, voiceprint recognition is still not good at distinguishing twins. A BBC reporter asked his non-identical twin brother to impersonate his voice and log into his HSBC account, which was successful for the eighth time.
HSBC later said in its response to BBC that the twins' voice prints were similar, but it was undeniable that the introduction of voice print recognition technology had greatly reduced fraud and had proved to be more secure than ordinary passwords.
In April this year, HSBC announced that it had been used by 1.6 million customers 1 million times since the introduction of voiceprint recognition in 2016, and that the system had prevented 330 million pounds worth of bank fraud.
In the future, voice prints may also help the police catch scammers who are difficult to track down.
In addition to the United Kingdom, some banks in the Netherlands, Macau and China have introduced voiceprint recognition systems one after another. As early as 2004, California-based BeepCard introduced a credit card that supports voice recognition, which can only operate properly after recognizing the user's voice.
In fact, the earliest large-scale application of voiceprint recognition technology was in the field of national defense and security. According to The Intercept, according to a classified memo issued by the National Security Agency (NSA) in January 2006, NSA analysts through a "voice recognition technology." A spy was identified in audio files and telephone surveillance.
The technology used in this is obviously voiceprint recognition, and because of the special status of NSA, it is possible to build a huge voiceprint database, which further improves the efficiency of recognition, even if the monitoring object uses an unknown number or a different language. The algorithms can match the data of the voiceprint library.
Nuance, a company that provides voiceprint recognition technology to military and intelligence agencies, says it can build a nationwide biometric voiceprint recognition system that can quickly and accurately identify a person's voice from millions of voiceprints. At present, the voiceprint recognition technology of Nuance has also been applied to bank transactions, car voice assistants and other fields.
Now similar voiceprint recognition technology has been applied to more consumer products, the most common of which are smart speakers, which are already supported by Amazon's Echo, Apple's HomePod and Ali's Tmall Elves. And WeChat, which is most commonly used by Chinese, actually launched a voice lock based on voiceprint recognition technology in 2015.
However, in the field of consumer electronics, most consumers do not have high acceptance of voiceprint recognition technology. For example, after four years of online WeChat voice locks, many users complain that the recognition rate is not high, and the security needs to be improved.
According to the Beijing News, misreading, misreading or playing the recording may open the WeChat sound lock, and then WeChat admitted that "there are certain hidden dangers."
This is also a sign that voiceprint recognition technology is not mature enough in the consumer field. Zheng Fang, director of the speech and language Technology Center of Tsinghua University, said in an interview with the media:
“At present, in the field of voiceprint recognition, different technology or product providers vary greatly; at the same time, voiceprint recognition is no longer a single emphasis on accuracy, but becomes more mature, complete and available. ”
If voiceprint recognition technology is popular in smartphones and many applications, it may also raise more questions about privacy. This means that it is possible for app to give us personalized recommendations by "eavesdropping" on our words.
A new Facebook patent last year allowed the phone's microphone to be activated remotely, recording the user's conversations and ambient sounds, and then sending the data back to Facebook for analysis.
Although there are many problems to be solved, it is undeniable that voiceprint recognition does have a wider application prospect. Biometric authentication technology has been considered to replace all kinds of plaintext passwords, and voiceprint has special advantages over other voiceprint features. In addition to being more secure and convenient, the cost of speech acquisition and recognition is also lower than that of facial recognition. In fact, it is more suitable for password authentication.
Gartner and other investigators have predicted that voice will replace the screen in the future, although it still seems far away, but the products represented by Amazon smart speakers have led to a growing share of the voice search market. Perhaps the day when voiceprint recognition becomes the mainstream authentication model will not be too far away.
|You may also want to read:|
|How to change the language of Xiaomi Mi band 4?|
|AMD Ryzen 7 3700X and Ryzen 9 3900X first benchmark of the in games released|
|AfterShokz releases Aeropex and Xtrainerz bone-conducting headphones|
● Over 300,000 products
● 20 different categories
● 15 local warehosues
● Multiple top brands
● Global payment options: Visa, MasterCard, American Express
● PayPal, Western Union and bank transfer are accepted
● Boleto Bancario via Ebanx (for Brazil)
● Unregistered air mail
● Registered air mail
● Priority line
● Expedited shipping
● 45 day money back guarantee
● 365 day free repair warranty
● 7 day Dead on Arrival guarantee (DOA)
2019-09-27By Goraud Mazanec
2020-03-09By Ethelind Lissy
2019-08-12By Sigismondo Eisenhower
2020-03-09By Pat Gottsdanker
2019-10-14By Adan Flannigan
2020-03-10By Bernadina Kempton