Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

The Case for Bi-modal

Нью-Йорк конференция
by

Елена Аронова

on 13 August 2014

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of The Case for Bi-modal

The Case for Bi-modal
Authentication

Bryan Chaney
Vice President
Sales & Marketing
SpeechPro USA
www.speechpro-usa.com

Security
Mobility
Customer
Experience
SpeechPro is the US subsidiary of the Speech Technology Center (STC)
STC is a World Leader in Voice Technology with customers in 70 countries
STC was established in 1990 & fully Certified ISO-9001:2008

STC has offices in US, Mexico, Germany, Finland with HQ in St. Petersburg Russia
“SpeechPro has established a leadership position in a fast-growing and potentially large sector of the voice biometrics marketplace”

Dan Miller,
Senior Analyst at USA
About SpeechPro
VOICEKEY EVOLUTION & SERVICE OFFERINGS
CUSTOMER
VOICEKEY OVERVIEW
MOBILE
APP
IVR
WEB
AGENT
CUSTOMER
VOICEKEY EVOLUTION & SERVICE OFFERINGS
VOICEKEY EVOLUTION & SERVICE OFFERINGS
Text Dependent &Text Independent:
Allows passphrase or normal speech
Liveness Detection:
Prevents replay attacks

High Accuracy:
Proprietary algorithms fusion (Formant, Pitch, TV)
Speaker Change Detection:
Prevents unauthenticated caller joining call
Cross-channel ID&V:
Allows multi - device authentication from voice print
Signal to Noise Ratio to 7dB
- High toleration of background noise
White-list:
VIP clients can be authenticated without passwords
Black-list:
Continuous search for known fraudsters in the stream of calls
OnePass: How does it work?
Enroll
OnePass: How does it work?
Collect
Process
Access
VOICEKEY.ONEPASS OVERVIEW
OnePass Architecture- One-Step Authentication
Authentication automatically launched from mobile app
Camera searches for user’s face

Audio recording launched with face detection
Static and dynamic passphrase simultaneously
Zero-tap process
OnePass Evolution —
put these tables in the arrow
Android
iOS
Current Q1-2014 Q2-2014 Q4-2014
One-tap authentication
Liveness detection, face&voice
Fused EER
Touchless initiation, app looks for face and self launches
Simultaneous static and dynamic passphrase
TBD
Mobile app captures face and voice Sample creation and comparison on server, face and voice
Mobile app captures face and voice, creates voice sample
Sample creation and comparison on server, face
and voice
Mobile app captures face and voice, creates face and voice sample. All comparisons
done on server
Current Q1-2014 Q2-2014 Q4-2014
TBD
One-tap authentication
Liveness detection, face&voice
Fused EER
Touchless initiation, app looks
or face and self launches
Simultaneous static and в
dynamic passphrase
TBD
Mobile app captures face and voice Sample creation and comparison on server, face and voice

Mobile app captures face and voice, creates voice sample
Sample creation and comparison on server, face
and voice
Mobile app captures face
and voice, creates face
and voice sample. All comparisons done
on server
The Case for Bimodal – Market Drivers
Security
Mobility
Customer
Experience

Voice Biometrics widely accepted, FA around 0.1%, FR under 2%
Face Biometrics becoming accepted, FA around 0.1, FR at 2.1
Fused biometrics: 0.01% FA with FR under 2%
Apple’s Touch ID, Siri, Facebook, cameras, Google Now, toys, brings familiarity, expectation
Explosion in malware, Android most targeted.
Also in iOS: iPhone users conduct mobile transactions more frequently than Android
Passwords “aren’t” anymore, must be changed, faster failure.
51% of consumers switched vendors due to a poor customer experience
75% used online channels, 33% of these used a mobile device.
82% feel companies they buy from cannot be trusted with PII and financial information

The Case for Bi-modal – Why Voice & Face?
Pro
Con
Voice Face Finger
Ubiquitous hardware
Friction of distance
Easy, familiar Proven
Ubiquitous hardware
Noisy environments
Widely deployed
No background noise issue Authentication time
Noisy environments (below 7dB SNR)
Enrollment time
Low light situations
Privacy concerns
Few devices with reader Liveness detection Leave it everywhere Political Correctness issues Friction of distance
Touchless Biometric Logins replaces traditional identification
Banking Corporate Paid Social Mobile Gaming
Apps Networks Content Media payments
False Acceptance (%)
False Rejection (%)
MULTI-MODAL APPROACH DRASTICALLY REDUCES EER
EER 0.5% or lower for multi-modal verification

EER 1.4-4.0% for face or voice alone
Mobile Web Physical Access Phone
CROSS-CHANNEL CAPABILITY
All Use One Single Voiceprint
User transition to mobile devices
for transactions is inevitable
No additional hardware required for either mode
EER’s lower
with two modes
Same voiceprint usable in different channels
AHT is reduced
in contact center
IVR containment
rate increased
Captured voiceprints
can become part of whitelist/blacklist
DB
SDK allows “under the covers” implementation
Simple licensing model, voiceprint licenses reusable
How will Bi-modal drives the bottom line?
The Case for Bi-modal
Authentication

Bryan Chaney
Vice President
Sales & Marketing
SpeechPro USA
www.speechpro-usa.com
Full transcript