In the Partner Portal you can find a compact overview of Voice Verify solution, advice and recommendations which build upon the basic information provided in Voice Verify webpage and via our communication.
Partner Portal also presents two ways of testing and getting hands on Voice Verify, which are easy-to-try and ready for you right now.
Phonexia Voice Verify is a cutting-edge voice verification solution designed specifically for contact centers to enhance the security of voice interactions whilst reducing call handling time and without compromising user experience.
To see all the benefits and use-cases of Voice Verify please visit Voice Verify webpage.
There are several possible ways to get hands on and test our technologies. The complete summary of all testing possibilities can be found here:
For demonstration purposes we created a Voice Verify DEMO. By using this demo you can get a feel for Voice Verify in a couple of minutes even without any programming or technical knowledge. This DEMO helps you understand the process of enrollment and verification and let’s you try our technologies on your own voice.
The only requirements for using our DEMO are:
Phonexia Voice Verify SANDBOX is a cloud appliance accessible via RESTful API that enables testing of our technologies on one’s own recordings. The SANDBOX allows clients to get a feel for the complete Voice Verify solution and simulates a real communication scenario between Voice Verify and another software.
Requirements for using our SANDBOX are:
The Sandbox is focused only on voice verification, in which case you expect a certain speaker on the line and ask the question “Is the speaker calling right now really the person you expect him to be?”.
To verify a speaker, Phonexia Voice Verify uses the power of voice biometry to recognize speakers automatically by their voice with extremely high accuracy.
Voice biometry is based on the fact that the speech organs and speech habits of every person are unique. As a result, the characteristics of the speech signal captured in a recording are also unique, thus the technology can be language-, accent-, text-, and channel-independent.
Phonexia’s voice biometry technology is called Speaker Identification. Speaker Identification is based entirely on deep neural networks called Phonexia Deep Embeddings™.
Speaker Identification extracts the features from a voice to a voiceprint. A voiceprint is a small numerical representation of the voice.
The whole voice verification process consists of two parts:
Voice biometry is (as well as finger or face biometry) never absolutely perfect.
Two types of mistakes can occur:
Based on these possible mistakes, accuracy can be measured. Phonexia’s Speaker Identification achieved a stunning 97% accuracy on call center datasets.
The accuracy can be boosted further with a calibration. Calibration adapts the Speaker Identification technology on a specific customer environment.
In the picture below you can see a basic technical flowchart describing interactions with Voice Verify.
Voice Verify is designed to process SIP (Session Initiation Protocol) based calls, HTTP streams or WebSocket streams as voice input. This means that it is a matter of configuring the Call Center’s PBX (Private Branch eXchange) or other audio source to fork each call and send a copy of the customer’s voice to Voice Verify.
After Voice Verify receives the voice, it is able to process it with Speaker Identification voice biometric technology. It can either enroll this voice into a database or verify againts already existing voiceprint from the database. This way it is able to verify customer’s identity.
Voice Verify provides result of verification of the consumer to the client’s Call Center software via RESTful API. The verification information needs to be incorporated into the client’s processes and software.
The latest technical documentation including hardware requirements can be found in the link below:
This section of the Partner Portal shall provide an overview of the pricing structure and discuss specifics of the pricing model. We understand that the future is hard to predict, therefore we created a volume-base tiered pricing model which will automatically adjust to your needs as you go.
For more pricing details and model examples please please visit Pricing.
To implement Voice Verify, internal processes of the customer have to be changed. This section describes things to think about before putting Voice Verify into production, what are the steps to production and how the PoC can be run and evaluated.
For more information about set-up and PoC process visit Proof of Concept process.
Voice Verify utilises third party components to ensure the best possible performance. The complete list of all third party tools for all Phonexia technologies and solutions can be found in a separate partner portal of Speech Platform for GOV. You can view the list here.
Thank you for choosing Phonexia Voice Verify and rest assured we will do our best to make you enjoy the benefits of the most advanced voice verification product on the market as soon as possible.
In case you have any additional questions, comments or feedback with regards to Voice Verify or to this document, feel free to reach out to any of your contacts in Phonexia s.r.o., visit www.phonexia.com or contact us at [email protected]. For technical support please use [email protected].