VoiceBiometrics
Voice

Voice biometrics

Is the safest and simplest way to integrate total security in voice apps – allowing the verification of the users’ identity in the normal flow of a conversation. It allows you not only to verify the identity of a user, but to identify which user it is.

1% Equal Error Rate (EER)

Multiple security alternatives fix text, variable text, free text

Mono Channel, multi-channel (channel compensation techniques)

Ability to work with 8 or 16 Khz

Anti-spoofing mechanisms

Proprietary Algorithms with multiple extensions adapted to each deployment and required characteristics.

Speech

Verbio’s speech recognition

Is one of the most advanced Continuous Speech recognition engines and can be used both in a telephone environment as well as in multimedia applications. Our speech recognition is optimized for conversational AI. Some of the main features include noise robust algorithms and advanced features like Blind Source Separation.

Telephony (8Khz) and Multimedia (16Khz)

+ 95% precission, depending on accoustic environment.

Broad language coverage

Deep Neural Networks (DNN) + HMM

Noise robustiness algorithms and training

VoiceBiometrics
text-to-speechvoices
Text

Verbio’s text-to-speech voices

Provide communication with a human touch by using expressivity. Our voices have been designed to be able to express intonations and emotions through a conversation.

Voices able to reproduce entonations and emotions.

Own combination between deep neural network and Hidden Markov Models

Broad language coverage

Ability to use a variety of languages at the same time.

Conversacional

Verbio's conversational sensors

Allow the extraction of important information from the conversation in order to analyse the dialog and improve communication between people and bots. Age detection, gender detection, and emotion detection are some of the sensors available.

More than 40 variables analyzed

Thousands of real customer audios analyzed

The sensors have been designed to add value to the business and allow decission making.

Integrated with the rest of the solutions

text-to-speechvoices
TheBrainModule
Brain

The Brain Module

is able to process information, understand intention, identify entities and provide an answer. The central brain stores all business information (rules, integrations, knowledge, user profiles and history, etc.) common in all channels.
Several services and media (Social Media, Dialog Manager, Spoken Dialog…) can be set up from the same core technology by using specific modules adapted to the specific channel.

Execute complex tasks

Progressively, the brain will be able to understand and perform more complex tasks that require more intelligence.

Invisible UI

As the intelligence of the NLP module improves, the need for a visual UI will disappear.

Context & Multilingual

Ability to understand, identify and extract contextual elements such as meaning in any language and extract the meaning of texts and social media.

Consistency

By having a central NLP brain, there is consistency in all our interactions.

Intelligence from unstructured information

The ability to extract information and meaning from complex repositories will provide very useful knowledge to the bots in their communication with humans..

Search just like you talk

By having a central NLP brain, we provide consistency in all our interactions.

Multiple intelligences

Unique approach by including multiple cognitive systems orchestrated through a central dialog manager.

On Premise or Cloud

Both Cloud or On-Premise Solution. Our technology can be deployed in any deployment architecture.

TheBrainModule