2023 Speech Industry Award Winner: ID R&D Pioneers Liveness Detection

Article Featured Image

The rapid advancement of speech synthesis technology and generative artificial intelligence has presented a real threat to consumers, spawning a vast array of fraud opportunities that would have been unthinkable just a few years ago. ID R&D, a New York-based provider of liveness detection and voice biometrics, has quickly become the one company that directly addresses these threats by combining passive facial liveness and voice anti-spoofing technologies, accurately identifying even advanced AI-based deceptions.

The company in just the past few months collected three very significant U.S. patents for technologies that do just that.

The first covers continuous biometric authentication in chatbots and virtual assistants on mobile devices. The method involves the collection of multiple biometrics, including voice, throughout the chat session to continuously confirm people’s identities.

The second sets the industry benchmark for preventing voice spoofs and deepfakes. The method combines an analysis of different speech audio features with deep convolutional neural networks to enhance detection accuracy. It applies a multi-stage approach and analyzes voice samples by first dividing them into frames, which enables rapid, real-time detection during conversational speech.

The final patent this year covers ID R&D’s novel approach to detecting the spoof attacks that threaten digital onboarding and biometric authentication. The method increases the speed and accuracy of IDLive Face, ID R&D’s product for frictionless facial liveness detection.

“True to our name, research is a strategic priority at ID R&D, and our products are built upon investigation and discovery of new ways to reduce fraud without burdening users,” said Alexey Khitrov, CEO and cofounder of ID R&D, in a statement.

The company’s recent success in ensuring the highest level of security against evolving voice threats while still delivering an imperceptible consumer experience follows its 2021 acquisition by Mitek Systems, a provider of digital identity verification.

“Today’s consumers are transacting in a digital-first world, and they expect to trust technology to keep them safe against burgeoning threats. Identity verification has never been more relevant,” said Max Carnecchia, CEO of Mitek, in a statement at the time. “Rapid advances in artificial intelligence are enabling novel forms of fraud, which means technologies like those pioneered by ID R&D are critical in this fight.”

The company, which was founded in 2016, continues to operate independently, though its technologies have been integrated into Mitek’s products.

That’s not the only integration that ID R&D has undergone in the past few months.

Tech5 integrated ID R&D’s IDVoice biometrics with its decentralized digital ID platform to enable tri-modal (face, fingerprint, and voice) biometrics and liveness detection.

Rahul Parthe, cofounder, chief technology officer, and chairman of TECH5, called ID R&D’s technology “top-ranked voice biometrics and liveness detection.”

iPulse Systems also integrated ID R&D’s IDVoice technology into its new voice biometrics-as-a-service solution hosted on the Microsoft Azure platform. Called VoiceIQ.cloud, it relies on IDVoice’s matching and liveness detection algorithms, which recently received an update to increase speed and precision.

Gary Chalmers, iPulse’s head, also sang ID R&D’s praises, saying the partnership “uses the ID R&D product to provide an industry-first solution that should prove completely disruptive in this sector, bringing the power of voice to anyone at an affordable price.”

And the company’s recognitions did not stop there. ID R&D claimed the top rank for speaker recognition with its IDVoice voice biometrics software in the VoxCeleb Speaker Recognition Challenge during last year’s Interspeech conference.

ID R&D’s IDVoice Verified voice verification and spoof detection product also took first place in the SASV 2022 Challenge for biometric speaker verification and liveness detection.

The SASV (which stands for spoofing-aware speaker verification) 2022 Challenge involved 23 submitters from vendors and academic institutions with advanced programs in voice biometrics.

And ID R&D’s emphasis on R&D continues.

“Rapid advancements in artificial intelligence bring new challenges to securing digital onboarding and authentication, particularly from deepfakes, and we expect this trend to persist,” Khitrov says. “Fortunately, we are also working hard to research and develop new ways to leverage AI to its full potential to counter AI-powered fraud.”

SpeechTek Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues
Related Articles

Safety and Ethical Concerns Loom Large in Voice Cloning

AI makes synthetic speech sound more realistic than ever—and therein lies the danger.