
Synthetic voices for audio verification research
Working with the RTVE-UGR Chair to develop synthetic voices that support research into detecting fake audio, helping fight disinformation.
“We have a research project or an AI idea and don't know how to turn it into something usable, and sellable.”
Services
We explore what's possible today in speech, language and conversation, test it with users and turn it into a prototype or pilot your organization can evaluate. We also join R&D projects as a technology partner.
Applied research takes an advance in voice, language or conversational AI all the way to a prototype tested with real people. We've done it since 2017 with universities such as URJC, UGR, UCLM and the University of Lleida, and in 2020 through a NEOTEC project, funded by Spain's CDTI, to build our own Spanish speech synthesis models.
We review open and commercial models, papers and tools, and tell you what works for your case, your language and your data.
We build fast to test the key hypotheses and questions: is it feasible, at what quality and at what cost.
We take the prototype into a real setting, measure and learn what matters most before more resources go in.
We work with universities, research chairs and companies on research projects, such as CDTI's NEOTEC program or the RTVE-UGR Chair.
We understand the challenge and review what today's technology can do for your case, your language and your data.
We build just enough to validate the idea with real people.
We try it out in your organization, measure and adjust.
We document and train your team so they can keep evolving it.
What you get
Who it's for

Working with the RTVE-UGR Chair to develop synthetic voices that support research into detecting fake audio, helping fight disinformation.

Synthetic voices to make real-time election coverage accessible for towns with fewer than 1,000 inhabitants, an AI project by RTVE.

Designing and building the platform for CyberFold, a strategic INCIBE project in which a Universidad Rey Juan Carlos research group applies manifold learning to network intrusion and anomaly detection.

A face-to-face translation mobile app so emergency medical staff can talk to patients who don't speak Spanish, without storing their voice.

Live translation so international students at a university school can follow classes taught in Spanish, in their own language, on their phones.
Ideas we're working on. If one fits a challenge of yours, we can pilot it together.
Find where your assistant gets lost, before your users do.
A platform to measure the quality of assistant and agent conversations, in Spanish and other languages.
Nobody should miss a class because of the language.
Real-time captions and translation so every student or attendee can follow the session in their own language.
Yes. We've worked with Universidad Rey Juan Carlos, Universidad de Granada, Universidad de Castilla-La Mancha and Universitat de Lleida, and with the RTVE-UGR Chair on synthetic voices for fake audio detection.
It depends on the question we want to answer. A proof of concept can take weeks; a research project, months.
With your organization's own budget or through public calls, where we take part as a technology partner: tell us about the call and the consortium and we'll see how we fit in. Our Spanish speech synthesis models came out of a NEOTEC project funded by Spain's CDTI in 2020, in collaboration with Rey Juan Carlos University and RTVE.
When the project allows it, yes. With the RTVE-UGR Chair we published a scientific paper on speech synthesis and news verification in Applied Sciences in 2024.
Custom synthetic voices for brands and media, in Spanish, its accents and Spain's co-official languages.
We design and build AI assistants and agents that converse naturally through voice, text and images, with a personality that fits your brand.
We measure how your assistant talks: what it understands, where it gets lost and when it shouldn't answer.
Conversation design, assistant personality, synthetic voice and conversational AI, taught by the people who build them.
Tell us what you want to achieve. If we're not the right team, we'll say so.