Solutions
Speech Technology
Products
Resources
Company
Contact
← All projects
← All projects

Innovation Project

Speech Technology Development

Speech Technology Development

Speech Technology Development

Optimizing speech processing through next-generation language model infrastructure

Optimizing speech processing through next-generation language model infrastructure

The QuickSpeech project aims to create a new research hub within VoiceInteraction that will research and develop new models, algorithms and software architectures to optimize language model processing.

Completed
Completed

QuickSpeech

-

Lisbon

1. Project Overview

QuickSpeech was created to address one of the most important challenges in modern speech technology: the increasing computational demands of language processing systems. The project established a dedicated research hub focused on developing new algorithms, architectures, and processing methodologies capable of improving the performance and scalability of speech and language technologies. By exploring hardware-agnostic approaches and advanced parallel processing techniques, QuickSpeech helps create a stronger technological foundation for future speech solutions.

2. VoiceInteraction's Contribution

VoiceInteraction led the development of technologies aimed at improving the efficiency and scalability of speech processing systems. The project focused on researching new approaches to language model processing and exploring architectures capable of supporting future speech technology requirements. Through QuickSpeech, VoiceInteraction strengthened its expertise in computational optimization, language processing infrastructure, and high-performance speech technology development.

3. Key Capabilities

  • Parallel language model processing

  • Hardware-agnostic speech infrastructures

  • GPU and FPGA optimization

  • Large-scale speech processing architectures

  • High-performance language model execution

  • Next-generation speech technology infrastructure

4. Outcomes & Impact

QuickSpeech contributes to the evolution of speech technology infrastructure by enabling faster language model processing and improved computational efficiency. The project supports future speech recognition solutions that can operate across different hardware environments, creating a more scalable foundation for next-generation speech products and AI applications.

5. Partners & Consortium

QuickSpeech was developed through collaboration between research, innovation, and technology stakeholders focused on advancing speech processing infrastructure and language model technologies.

The project combines expertise in speech processing, hardware optimization, and computational efficiency to support the development of next-generation speech technology solutions.

6. Project Information

Project Code

LISBOA-01-0247-FEDER-042679

Funding Programme

Lisbon 2020, Portugal 2020, FEDER

Intervention Region

Lisbon

Beneficiary Entity

VOICEINTERACTION - TECNOLOGIAS DE PROCESSAMENTO DA FALA S.A.

7. Funding Information

Main Objective: Strengthen research, technological development and innovation. Approval date: 2019-03-28. Start date: 01-02-2019. Completion date: 2021-01-31. Total eligible cost: €263.471,87. Financial support from the European Union | FEDER: €105.388,75. Support logos: Lisbon 2020, Portugal 2020, FEDER.

← Back to all projects

CONTINUE READING

Related projects

Explore other research and innovation initiatives connected to speech technology, AI, and operational workflows.

Explore other research and innovation initiatives connected to speech technology, AI, and operational workflows.

RHESIS

RHESIS

Speech Technology Development
Speech Technology Development

Robust Privacy-Preserving Adaptive Human-AI Dialogue Systems for Defence Applications

Robust Privacy-Preserving Adaptive Human-AI Dialogue Systems for Defence Applications

Developing trustworthy human-AI dialogue systems that allow defence and security personnel to interact naturally with AI through voice and text, improving access to information, operational decision support, and multilingual collaboration across European defence environments.

View Project →
View Project →

LINGUARISE-DC

LINGUARISE-DC

Speech Technology Development
Speech Technology Development

Language Innovations: Nimble, Guided, User-friendly, Agile and Robust Solutions Eliminating Defence Challenges

Language Innovations: Nimble, Guided, User-friendly, Agile and Robust Solutions Eliminating Defence Challenges

The LINGUARISE-DC project develops an advanced language platform to address linguistic challenges in Europe’s defence sector. It enhances EU missions by improving communication, intelligence gathering, and operations, filling a key gap in Human Language Technologies.

View Project →
View Project →

QREN 5108

QREN 5108

Product Innovation
Product Innovation

Enhancing multimedia intelligence through semantic information management

Enhancing multimedia intelligence through semantic information management

This project is integrated in the distribution and filtering of semantic information associated with multimedia contents.

View Project →
View Project →

QREN 2525

QREN 2525

Product Innovation
Product Innovation

Creating natural dialogue systems for automated customer interaction

Creating natural dialogue systems for automated customer interaction

The Dialogue Management System will allow, in broad terms, a natural conversation between a user, client or consumer and a given entity, without the intervention of a human operator.

View Project →
View Project →

SAVAS

SAVAS

Multilingual Technology
Multilingual Technology

Building multilingual resources for speech and language technologies

Building multilingual resources for speech and language technologies

The SAVAS project collected textual and audiovisual resources produced in six European languages: Spanish, Basque, Portuguese, Italian, French and German.

View Project →
View Project →

CAPER

CAPER

Law Enforcement & Security
Law Enforcement & Security

Enabling collaborative intelligence against organized crime

Enabling collaborative intelligence against organized crime

The CAPER project aims to prevent organized crime through the creation of a common platform that facilitates sharing, exploring and analyzing publicly accessible information, as well as information from private repositories.

View Project →
View Project →

AGATHA

AGATHA

Law Enforcement & Security
Law Enforcement & Security

Intelligent analysis system in open sources of information for surveillance/crime control

Intelligent analysis system in open sources of information for surveillance/crime control

The AGATHA consortium aims to develop a system for assisting criminal investigation police and intelligence services to facilitate the compilation of evidence of criminal practices by using information available in open sources and analyzing it automatically.

View Project →
View Project →

CallScriber

CallScriber

Law Enforcement & Security
Law Enforcement & Security

Transforming intercepted call analysis with automated transcription

Transforming intercepted call analysis with automated transcription

This project aims to create an integrated platform where the contents of intercepted calls are automatically transcribed using automatic speech recognition technology and segmented according to the speaker.

View Project →
View Project →

VOISCRIPTUM

VOISCRIPTUM

Operational Workflows
Operational Workflows

Automating judicial transcription and dictation workflows

Automating judicial transcription and dictation workflows

The main goal of this project is to automate transcriptions of interrogations and depositions for the purpose of maximizing resources and reducing the current burden and time spent in manual transcription procedures from scratch.

View Project →
View Project →

QuickSpeech

QuickSpeech

Speech Technology Development
Speech Technology Development

Optimizing speech processing through next-generation language model infrastructure

Optimizing speech processing through next-generation language model infrastructure

The QuickSpeech project aims to create a new research hub within VoiceInteraction that will research and develop new models, algorithms and software architectures to optimize language model processing.

View Project →
View Project →

Past Internationalization Plan

Past Internationalization Plan

International Growth
International Growth

Expanding VoiceInteraction’s presence in international markets

Expanding VoiceInteraction’s presence in international markets

Internationalization plan supporting business growth, international market development and export expansion.

View Project →
View Project →

VoiceInteraction’s Internationalization

VoiceInteraction’s Internationalization

International Growth
International Growth

Expanding speech technology solutions into global markets

Expanding speech technology solutions into global markets

With this project, VoiceInteraction intends to strengthen its internal framework for the competitive capacity necessary for internationalization, in order to ensure its continuous growth across borders.

View Project →
View Project →

AIDA

AIDA

Law Enforcement & Security
Law Enforcement & Security

Applying artificial intelligence to cybercrime and terrorism investigations

Applying artificial intelligence to cybercrime and terrorism investigations

The AIDA Project aims to create a platform for compiling, processing and visualizing data in order to automate the investigation work of law enforcement entities in the participating states in the fight against cybercrime and terrorism.

View Project →
View Project →

Qualification / Certification Project

Qualification / Certification Project

International Growth
International Growth

Strengthening digital transformation and international competitiveness

Strengthening digital transformation and international competitiveness

The Qualification Project aims to prepare the company for the transformation of its business model, intensifying its internationalization, diversifying geographic markets, market segments and customers, focusing more on digitalization, e-commerce and Industry 4.0.

View Project →
View Project →

HOUDINI

HOUDINI

Product Innovation
Product Innovation

Creating immersive cultural heritage experiences with speech interaction

Creating immersive cultural heritage experiences with speech interaction

This project seeks to stimulate the Tourism Industry through immersive and interactive experiences, powered by augmented reality technology and AI-based speech interaction.

View Project →
View Project →

PICCH

PICCH

Cultural Heritage
Cultural Heritage

Unlocking colonial audiovisual heritage through language technologies

Unlocking colonial audiovisual heritage through language technologies

This International Research Project aims to use language technologies to aid Archives and Researchers to better approach colonial audio-visual heritage and understanding of language shifts for an improved keyword search.

View Project →
View Project →

STARLIGHT

STARLIGHT

Law Enforcement & Security
Law Enforcement & Security

Building resilient AI capabilities for European law enforcement

Building resilient AI capabilities for European law enforcement

Starlight is an international research project aiming to support Law Enforcement agencies in cybersecurity operations against adversarial Artificial Intelligence.

View Project →
View Project →

EMPOWER

EMPOWER

Speech Technology Development
Speech Technology Development

Developing speech and audio intelligence for real-world applications

Developing speech and audio intelligence for real-world applications

VoiceInteraction participates in European research initiatives that advance speech intelligence, multilingual AI, operational automation, and trustworthy AI systems. Through collaboration with industry, academia, and public organizations, we help transform research outcomes into deployable technologies.

View Project →
View Project →

Operational speech workflows require different approaches

Discuss transcription, monitoring, accessibility, or conversational analysis requirements with the VoiceInteraction team.

ON THIS PAGE