GOVERNMENT & PUBLIC ADMINISTRATION INFRASTRUCTURE
Modernizing Legislative Transparency: AI-Supported Multilingual Transcription at the Spanish National Parliament
Parliamentary language is not simply recorded. It becomes law, archive, and public memory. As Spain expanded the use of co-official languages in the Congreso de los Diputados, transcription moved from administrative function to institutional infrastructure.
In this context, the Spanish Parliament integrated AI-supported transcription services to manage multilingual legislative discourse at national scale. The initiative reflects a broader shift toward domain-adapted AI systems designed for environments where precision, traceability, and linguistic diversity carry legal and political weight.
Building on a framework previously deployed at the Valencian Parliament, Pangeanic extends its human-enhanced transcription architecture to Spain’s primary legislative body, supporting the transition from manual workflows to a governed, AI-assisted model.
Task
AI-supported transcription to support multilingual Parliamentary integrity in a plurilingual state
The Spanish Parliament required a robust and scalable infrastructure to transcribe plenary sessions, committee meetings, and official proceedings across a multilingual environment that now includes Spanish, Catalan, Galician, and Basque.
The objective was threefold:
First, to ensure that the Diario de Sesiones, the official parliamentary record, maintains its legal and historical integrity across languages, without loss of nuance or procedural meaning.
Second, to reduce the latency between debate and publication, improving the usability of parliamentary output for internal teams, media, and public audiences.
Third, to expand accessibility in line with European directives on open data and digital public services, ensuring that parliamentary activity remains intelligible to a linguistically diverse society.
This effort forms part of a broader linguistic services framework, within which AI-supported transcription contributes to the modernization of documentation workflows.
Aligned with the European MOSAIC Media Initiative
The Spanish Parliament project benefits from Pangeanic’s active role in the European MOSAIC Project, a funded program that drives innovation in cross-lingual, AI-assisted multimedia accessibility. Knowledge transfer from MOSAIC’s research outcomes has been instrumental in building a methodological framework for public-sector transcription and automatic subtitling workflows across Europe.
Through these synergies, Pangeanic shows how EU-funded R&D meets real-world institutional needs, creating lasting value for multilingual governance and accessibility.
Pangeanic's solution
Domain-Adapted AI Workflows for Institutional Governance
Rather than relying on general-purpose models, the system is structured around domain-adapted AI workflows designed for parliamentary language. This reflects a growing recognition across enterprise and public-sector environments that generic systems struggle in contexts where terminology, procedure, and multilingual nuance intersect.
The result is not a replacement of human expertise, but a redistribution of effort: automation accelerates the first layer of transcription, while linguistic specialists ensure that outputs meet institutional standards.
Core Capabilities
Specialized Speech Recognition
Models adapted to parliamentary speech patterns, including procedural language, speaker variation, and technical political vocabulary, improve the fidelity of initial transcription.
Human-in-the-Loop Validation
Expert linguists review, correct, and validate outputs, ensuring that the final record aligns with the authoritative tone and precision required in legislative documentation.
Integrated Multilingual Processing
A unified workflow supports Spanish, Catalan, Galician, and Basque within the same operational environment, treating regional languages with equal technical rigor.
Governed Infrastructure
The system operates within a controlled environment aligned with public-sector requirements for data protection, traceability, and operational oversight.
R&D Continuity
Methodological foundations draw on prior European research initiatives, including the EU-funded MOSAIC Project, where multilingual media and accessibility workflows were developed and validated in public-sector contexts.
Analysis and Impact
Operational Efficiency and Democratic Inclusion
The adoption of AI-supported transcription introduces a structural improvement in how parliamentary language is processed, stored, and accessed.
Several effects emerge.
Enhanced operational precision
Human expertise shifts from manual transcription to higher-value review and validation, improving both efficiency and consistency.
Faster access to parliamentary content
Reduced turnaround times allow debates to become searchable and reusable within significantly shorter timeframes.
Improved accessibility
Multilingual transcription and subtitling support broader participation by citizens, media, and institutional stakeholders across Spain’s linguistic communities.
Strengthened linguistic representation
Basque, Galician, and Catalan are handled within a system that preserves their institutional role in parliamentary discourse, reinforcing their practical use in legislative activity.
Regulatory alignment
Workflows support compliance with Spain’s transparency requirements and the EU Accessibility Act, where access and clarity are not optional but expected.
Analysis and Impact
Operational Efficiency and Democratic Inclusion
The adoption of AI-supported transcription introduces a structural improvement in how parliamentary language is processed, stored, and accessed:
Enhanced operational precision
Human expertise shifts from manual transcription to higher-value review and validation, improving both efficiency and consistency.
Faster access to parliamentary content
Reduced turnaround times allow debates to become searchable and reusable within significantly shorter timeframes.
Improved accessibility
Multilingual transcription and subtitling support broader participation by citizens, media, and institutional stakeholders across Spain’s linguistic communities.
Strengthened linguistic representation
Basque, Galician, and Catalan are handled within a system that preserves their institutional role in parliamentary discourse, reinforcing their practical use in legislative activity.
Regulatory alignment
Workflows support compliance with Spain’s transparency requirements and the EU Accessibility Act, where access and clarity are not optional but expected.
“This implementation reflects a broader transition toward specialized AI applications in governance.” said Manuel Herranz, CEO of Pangeanic. “By integrating high-precision technology into parliamentary workflows, the objective is not only efficiency, but ensuring that documentation remains aligned with the linguistic and institutional requirements of a modern democracy..”
These results reinforce Pangeanic’s position as a trusted technology provider for parliamentary transcription and multilingual accessibility initiatives around the world.
About Pangeanic
Pangeanic develops multilingual AI systems, language technology, and AI data operations for public institutions and enterprises. Recognized with the Innovative SME seal by the Spanish Ministry of Science and Innovation, the company focuses on domain-adapted AI workflows where precision, privacy, and multilingual capability are essential.
From public administration to international organizations, Pangeanic combines linguistic expertise with controlled AI deployment to improve how complex information is processed, accessed, and governed.
Nos gusta la filosofía de trabajo de Pangeanic así como su profesionalidad. Su constante trabajo de escucha activa hacia el cliente les hace mejorar cada día más y esa, bajo mi punto de vista, es una de sus grandes cualidades.
Servicio rápido y eficiente. Encontramos la creación de una base de datos de traducciones realizadas por parte de Pangeanic una herramienta muy útil para otras ocasiones y/o trabajos similares.
La calidad es excelente, como de costumbre. El texto de origen cambió muchas veces durante el proceso de traducción. Pangeanic reaccionó rápidamente a los cambios, lo cual nos ayudó mucho.
Would you like to become our next success story?
Share your use case and security requirements. We’ll recommend the best AI deployment model (private SaaS, on-prem, or air-gapped), integrate it with your systems, and build a solution you can operationalize with confidence.
