SUMMA · Project

Multilingual Media Monitoring Platform That Tracks Global News Across Languages Automatically

digitalPilotedTRL 7

Imagine you need to keep tabs on what every TV channel, radio station, and news website in the world is saying — but they're all speaking different languages. SUMMA built a system that listens to all of these at once, translates everything on the fly, spots the same story popping up across countries, and gives you a clean summary. Think of it like a universal news radar that works in any language. It was actually tested live at the BBC and Deutsche Welle, two of the biggest broadcasters in Europe.

By the numbers

consortium partners across multiple sectors

countries represented in the consortium

total project deliverables produced

validated use cases (external monitoring, internal monitoring, data journalism)

core technical objectives achieved

major broadcasters used as field-test sites (BBC, Deutsche Welle)

The business problem

What needed solving

Global media monitoring has become nearly impossible to do manually. The number of broadcast channels, online news sources, and social media streams has exploded — and they publish in dozens of languages. Companies and newsrooms that need to track what's being said about them, their industry, or emerging crises are drowning in multilingual data they can't process fast enough.

The solution

What was built

SUMMA built a scalable media monitoring platform that combines automatic speech recognition, machine translation, story clustering, entity extraction, sentiment detection, and summarization into one pipeline. The project delivered 38 deliverables including a beta prototype and final version of external and internal media monitoring demonstrators, field-tested at BBC and Deutsche Welle.

Audience

Who needs this

International news agencies and broadcasters monitoring global media feedsCorporate PR and communications teams tracking brand mentions across multiple marketsGovernment intelligence and policy units monitoring cross-border media narrativesFinancial services firms doing multilingual media sentiment analysis for market signalsMedia analytics companies looking to add cross-lingual capabilities to their products

Business applications

Who can put this to work

Media & Broadcasting

enterprise

Target: International news agencies and broadcasters

If you are a global broadcaster dealing with the impossible task of monitoring news in dozens of languages — this project developed a scalable media monitoring platform, field-tested at BBC and Deutsche Welle, that automatically transcribes, translates, and clusters stories across languages. It handles speech recognition, machine translation, and story identification in one pipeline, turning a multilingual flood into actionable intelligence.

Corporate Communications & PR

mid-size

Target: PR agencies and corporate communications departments

If you are a PR firm or corporate comms team struggling to track how your brand is covered across 5 countries and multiple languages — this project built tools for entity extraction, sentiment detection, and cross-lingual summarization. Instead of hiring native speakers for each market, the platform aggregates mentions and detects emerging story-lines automatically across broadcast and internet media.

Government & Intelligence

enterprise

Target: Government agencies and policy monitoring units

If you are a government agency that needs to understand media narratives forming across borders — this project created an extensible knowledge base construction system that extracts entities, relationships, and sentiments from multilingual media streams. The platform was built with 8 partners across 5 countries and designed specifically to cope with the massive scale of global media data.

Frequently asked

Quick answers

What would it cost to deploy this kind of multilingual monitoring system?

The project's EU contribution is not available in the dataset, so exact development costs cannot be stated. However, the platform was built by a consortium of 8 partners including 2 SMEs and field-tested at BBC and Deutsche Welle, indicating enterprise-grade capability. Licensing or deployment costs would need to be discussed directly with the consortium.

Can this handle the volume of media we monitor (thousands of sources per day)?

Scalability was the core design goal — the project name literally stands for 'Scalable Understanding of Multilingual Media.' The platform was built to handle the global news monitoring problem across broadcast and internet channels in many languages simultaneously. It was stress-tested in real newsroom conditions at BBC and Deutsche Welle.

Who owns the IP and can we license the technology?

The project was coordinated by the University of Edinburgh under an RIA (Research and Innovation Action) funding scheme. IP is typically shared among the 8 consortium partners across 5 countries. Licensing discussions should be directed to the coordinator or specific technology partners like the 2 SMEs in the consortium.

What languages does the system support?

The project focused specifically on multilingual and cross-lingual capabilities as one of its 6 core objectives. While the exact list of supported languages isn't specified in available data, the consortium spans 5 countries (CH, DE, LV, PT, UK) and the platform includes speech recognition and machine translation components designed to be extensible to new languages.

How mature is the technology — is it ready for production use?

The project delivered both a beta-version prototype and a final version of its external media monitoring demonstrator, plus an internal monitoring demonstrator. These were field-tested at BBC and Deutsche Welle, and further validated through innovation events like BBC NewsHack. This puts the technology well past prototype stage into piloted territory.

Does it comply with media industry regulations and data handling requirements?

Based on available project data, specific regulatory compliance details are not documented in the project objectives or deliverable descriptions. However, field-testing at BBC and Deutsche Welle — both public service broadcasters with strict editorial and data standards — suggests the platform was designed with professional media requirements in mind.

Can we integrate this with our existing newsroom or monitoring tools?

The platform was explicitly designed to be 'sustainable and maintainable' with extensibility as a core objective. It provides rich visualizations based on multiple views and handles many data streams. The 38 total deliverables suggest comprehensive documentation and API specifications that would support integration work.

Consortium

Who built it

The SUMMA consortium brings together 8 partners from 5 countries (Switzerland, Germany, Latvia, Portugal, UK), with a mix that includes 3 universities, 2 industry players, 1 research organization, and 2 other entities. The 25% industry ratio and 2 SMEs suggest this was primarily a research-driven project, but the presence of BBC and Deutsche Welle as field-testing partners adds serious commercial credibility. The University of Edinburgh, a world leader in natural language processing and machine learning, led the coordination. The geographic spread across Western and Eastern Europe indicates the multilingual capabilities were tested across genuinely diverse language families.

THE UNIVERSITY OF EDINBURGHCoordinator · UK
DEUTSCHE WELLEparticipant · DE
THE UNIVERSITY OF SHEFFIELDparticipant · UK
FONDATION DE L'INSTITUT DE RECHERCHE IDIAPparticipant · CH
BRITISH BROADCASTING CORPORATIONparticipant · UK
UNIVERSITY COLLEGE LONDONparticipant · UK
PRIBERAM INFORMATICA SAparticipant · PT

How to reach the team

The coordinator is the University of Edinburgh (UK). SciTransfer can help arrange an introduction to the project team and relevant technology partners.

Order a report on this consortium See THE UNIVERSITY OF EDINBURGH profile →

Next steps

Talk to the team behind this work.

Want to explore how SUMMA's multilingual monitoring technology could work for your organization? SciTransfer can connect you with the right people in the consortium and help you evaluate fit. Contact us for a one-page technology brief.

Order a report on this topic More in Digital & ICT