Created on
ScreenReaderLB
- Call for solution type:
- SpeedUP
- participation dates:
- from 31-03-2025 to 30-04-2025
- StatuS:
- Open for applications
- Deadline for QUESTIONS:
- 21-04-2025
- Optional information meeting:
- 23-04-2025
- CONTACT:
- accessibilite@sip.etat.lu
Name
ScreenReaderLB
Ordering party
Information and Press Service (SIP)
Type
Software and its documentation
Objectives
Everyday an increasing amount of content is made available on the web in Luxembourgish. Unfortunately, this content is less accessible to people with disabilities due to the lack of appropriate assistive technologies.
Users who are blind use a screen reader software to interact with a computer or smartphone. The screen reader can display text via a Braille display and text-to-speech.
Due to the lack of text-to-speech technology in Luxembourgish, users who are blind or who have a severe visual impairment will either read text in Luxembourgish using a German text-to-speech or opt for a version of the same text in another language. Text-to-speech can also be useful for other situations, such as offering assistance to dyslexic people who may find reading difficult.
In this project, the Information and Press Service proposes to develop a new speech synthesiser for the Luxembourgish language, that can be used with several screen readers.
Challenge details
The main goal of this project is to produce an offline and cross-platform text-to-speech engine which will support the following combinations of OSes and screen readers: Jaws and NVDA on Windows, Talkback on Android, VoiceOver on iOS and MacOS.
These developments would be open source to enable easy integration with the current screen readers on the market for any citizen, without requiring advanced technical skills or specific hardware.
To protect personal data, the voice synthesiser must be able to operate locally, i.e. directly on the hardware of the user.
Particular attention will be paid to the quality and responsiveness of the solution so that it can be used in the context of interaction with a screen reader and a high speech rate.
The construction of such a speech synthesizer would be based on a text-to-speech model or raw data that would be provided by the Centre for the Luxembourgish Language (Zenter fir d’Lëtzebuerger Sprooch - ZLS) under licensing terms that allow this re-use.
This solution would have to be tested by the target audience, and these tests would make it possible to assess the suitability of the proposed solution for their needs. These tests would be carried out in partnership with the Centre for skills vision development (Centre pour le développement des compétences relatives à la vue - CDV).
For complete information regarding the context to be considered, the requirements, the deliverables and the type of offer to be submitted, please refer to the specifications annexed to this document.