SpeechPro: Russian biometric software capable of storing, identifying millions of voice samples

Madison Ruppert, Contributor
Activist Post

The use of biometrics and government-run centralized biometric databases is on rise and it seems like every day brings a new identification method including pedo-biometrics (using feet to identify targets), remote biometrics (using surveillance cameras), soft biometrics (which can be deployed on drone platforms), iris scans (which people are being illegally pressured into submitting to), high-speed facial recognition software (the use of which is being expanded to police departments across the United States by the FBI) and even so-called behavioral recognition software.

According to the FBI’s Biometric Center of Excellence, voice recognition is a “popular choice for remote authentication due to the availability of devices for collecting speech samples (e.g., telephone network and computer microphones) and its ease of integration, speaker recognition is different from some other biometric methods in that speech samples are captured dynamically or over a period of time, such as a few seconds.”

We’ve also seen strange implementations of voice recognition technology as evidenced by the deployment of voice recognition avatars at border crossings and even voice recognition technology in police cars.

Now the Russian Speech Technology Center, which, according to Slate, operates as SpeechPro in the United States, has created a program called VoiceGrid Nation capable of storing and identifying massive numbers of voice samples for governments around the world.

The software, at least according to the company behind it, is incredibly fast. It can deal with a database containing millions of voice samples of regular people, criminals, persons of interest or people on a watch list.

Computerworld reports that VoiceGrid uses three methods for voice matching along with an algorithm that automatically compares “voice models against voice recording obtained from different sources such as cell phones, land lines, covert recordings and recorded investigative interviews.”

When combined, VoiceGrid is capable of 90% accuracy within just 15 seconds.

According to Homeland Security News Wire, “Officials at VoiceGrid say that to get a sample, it only takes three seconds of a speech pattern to use for analysis. In five seconds it can search through and match 10,000 voice samples, executes up to 100 simultaneous searches, and stores up to 2,000,000 samples.”

According to SpeechPro, the accuracy is at least 90 percent and has already been deployed to Mexico and, according to SpeechPro’s president Alexey Khitrov, they are also working with multiple U.S. state and federal agencies.

“He declined to reveal any names because of nondisclosure and confidentiality agreements,” writes Ryan Gallagher for Slate. “But Khitrov did divulge that various versions of the company’s biometric technology are used in more than 70 countries and that the Americas, Europe, and Asia are its key markets.”

SpeechPro doesn’t just design voice recognition software for law enforcement and governments. They also have created technology for call centers which can verify the identity of customers automatically.

According to Agentura, a Russian secret services watchdog, the Speech Technology Center’s products have been sold off to several questionable governments including Belarus, Kazakhstan, Thailand and Uzbekistan.

Gallagher points out that this is hardly comforting given the incredible power this technology could give to an authoritarian regime.

“It has the technical capacity, for example, to store a voice-print of every single citizen in a country the size of Bahrain—with a population of 1.3 million—which would allow state security agencies to very effectively monitor and identify phone calls made by targeted political dissidents (or anyone else for that matter),” Gallagher rightly states.

Khitrov attempted to brush away these legitimate and major concerns by saying, “We just make sure that we work with trusted law enforcement agencies and try to make sure that they use it properly.”

Khitrov laughably claims that SpeechPro’s technology is solely used for “very noble causes,” although he was only able to cite a single example in Mexico where it was used to identify and find kidnappers who made ransom calls soon before they were going to murder someone.

To prove just how absurd Khitrov’s claim is, when Gallagher pressed for more examples of how VoiceGrid is being used in Mexico, Khitrov was forced to admit, “We don’t know the specifics because that’s their information.”

In other words, they actually have no clue about how it is being used, they just have a few nice anecdotes which make it seem like it’s only being used for noble ends.

Please support our work and help us start to pay contributors by doing your shopping through our Amazon link or check out some must-have products at our store.

This article first appeared at End the Lie.