This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 778450. Project coordinator: Verbio Technologies, Spain.


Verbio consists of several modules that can work together or can be integrated as single modules into voice recognition software: CSR (Continuous Speech Recognition) including Voice Biometrics, NLU (Natural Language Understanding), Speech Analytics (SA) and Text-to-Speech (TTS). Verbio’s technologies can be deployed in a variety of scenarios: embedded, cloud-based or hosted on clients own server.

Continuous speech recognition

Verbio’s speech recognition is one of the most advanced Continuous Speech recognition engines, allows users to speak in free form to a device and receive real time transcription without any delay. Verbio speech recognition is optimized for various acoustic scenarios, Real time transcription can be viewed and edited by the user and/or fed to a natural language engine for subsequent processing.

Voice biometrics


Verbio Biometric Solution allows customers to be verified without any specific enrolment process and identifying the user by their conversation.

  • Verbio solution can be used for authentication or identification purposes.
  • Main Characteristics for Verbio’s biometric solution: Language independent, < 9 seconds for training, < 3 seconds for verification, EER:0,3% (depending on conditions).

Natural Language Understanding


Our Cognitive Engine is designed to answer knowledge questions from users. Our virtual agent bases its responses on Ontology and Language statistics. Ontology is a known computational resource that formally expresses knowledge about a specific domain by means of a graph, creating objects, relations and properties. Each entity has a pattern that will identify all its name variants in open text, sentences are searched for with all entity patterns, overlapping entities are disambiguated with ontological rules.

Speech Analytics

Verbio Speech & Text Analytics solution extracts information from the dialogues, users and service processes, to analyze and show relevant conclusions about what has been said, by whom and how: Real Time & Batch Analysis, Multiple sensors (Age detection, Gender detection, Speakers Overlapping, Emotional detection), Main features are highly configurable allowing to easily define what needs to be

Text-to-Speech (TTS)

Verbio’s technology for TTS allows us to create expressive human-like voices with a fraction of the traditional cost in less than 2 months. Verbio’s technology for TTS is focused on providing a human experience through prosody and vocal quality
Verbio has also developed several custom voices for customers: ANSES, BBVA, PELAYO, VOX.

Visit us at

Share this