Jump to content

Wikispeech/id

Add links
From Meta, a Wikimedia project coordination wiki
This page is a translated version of the page Wikispeech and the translation is 16% complete.
Other languages:

Welcome to Wikispeech!

"Bayangkan sebuah dunia di mana setiap orang dapat saling berbagi segala pengetahuan dengan bebas."

This is the vision of the Wikimedia Movement. But today, access to knowledge is still largely dependent on the ability to read.

What is Wikispeech and why does it matter?

Wikispeech is an open-source text-to-speech solution built for Wikipedia and other Wikimedia platforms. It makes it possible to take in articles via audio—opening up free knowledge in a fundamentally new way.

Unlike most existing text-to-speech solutions, which are often costly and proprietary, Wikispeech is being developed as an open and accessible alternative, designed for people who prefer to take in text by listening.

What makes Wikispeech unique is not just that it converts text to speech. It is designed to be:

  • Open and community-driven – built and improved by volunteers, like Wikipedia itself
  • Accessible by default – no expensive software or advanced devices required
  • Scalable across languages – with tools to collect speech data and develop new voices over time

In short: Wikispeech is not just a feature. It is infrastructure for making knowledge accessible beyond text.

Why this matters

Today, hundreds of millions of people, and about 10% of the world's population, are excluded from written knowledge. Around 775 million people worldwide are illiterate. Nearly 300 million live with visual impairments, and millions more face cognitive challenges such as dyslexia, ADHD, or aphasia.[1][2][3]

As long as knowledge is primarily text-based, large groups remain excluded.

Wikispeech aims to change that by making knowledge accessible through listening. It is designed for people who cannot easily access written content but benefits anyone who prefers to learn by listening.[4]

By expanding how knowledge can be accessed, Wikispeech contributes to Wikimedia’s goal of knowledge equity: ensuring that more people, regardless of ability, language, or context, can take part in and benefit from free knowledge.

Current focus

During the second half of 2026 and the beginning of 2027, the project focuses on the development of a speech data collector. The software developed is based on the tool Lingua Libre with adaptations to the Swedish context and needs. The data will mainly be collected through crowdsourcing and will help develop new language technology tools in Swedish, refine existing language tools and contribute to research on dialects and speech impairments.

To collect data, we will collaborate closely with our partner organisations, the Swedish National Association for Aphasia, Begripsam AB, the Swedish National Association for Dyslexia and Dyscalculia, FUB (the Swedish National Association for People with Intellectual Disabilities), KTH Royal Institute of Technology, the Swedish Agency for Accessible Media, STTS (Södermalms talteknologiservice AB) and of course, the Wikimedia community. This part of the project will consist of events and activities where volunteers will be able to record their voice and share the data anonymously on Wikimedia Commons.

An analysis of how to collect the data in accordance with the EU’s GDPR has been conducted in order to make the collection process lawful and to understand which practice suits the purpose best. The information about the process, how the data will be handled and what volunteers consent to will be openly available and clearly stated during the collection process.

The project focuses on making Wikipedia content more accessible for people with cognitive disabilities. This work is carried out in collaboration with our Swedish partner organizations:

  • The Swedish National Association for Aphasia
  • Begripsam AB
  • The Swedish National Association for Dyslexia and Dyscalculia
  • FUB (the Swedish National Association for People with Intellectual Disabilities)
  • KTH Royal Institute of Technology
  • MTM (the Swedish Agency for Accessible Media)
  • STTS (Södermalms talteknologiservice AB)

Together, we explore how content, structure, and audio can better support different ways of understanding information.

Built with and for the community

In true Wikimedia spirit, Wikispeech is developed as an open and collaborative project. Volunteers can:

  • improve pronunciation and voice quality
  • identify and correct errors
  • contribute to new languages and voices

We are also developing tools to collect speech data, enabling continuous improvement and expansion.

All tools and data are released under free licenses, making them useful not only within Wikimedia, but also for the broader ecosystem of open speech technology.

Looking ahead

Wikispeech aims to become available across Wikimedia platforms, allowing anyone to listen to knowledge without needing special tools or advanced technology.

By making knowledge accessible in more ways, we move closer to a world where everyone can truly take part in the sum of all knowledge.

Referensi

  1. ↑ Kontributor Wikipedia. (19 September 2020). List of countries by literacy rate. Di Wikipedia, Ensiklopedia Bebas. Diambil 14:56, 6 Oktober 2020, dari https://en.wikipedia.org/w/index.php?title=List_of_countries_by_literacy_rate&oldid=979289293
  2. ↑ Pascolini D, Mariotti SPM. Global estimates of visual impairment: 2010. British Journal Ophthalmology Online. Pertama dipublikasikan pada 1 Desember 2011 di 10.1136/bjophthalmol-2011-300539.
  3. ↑ Max Roser dan Esteban Ortiz-Ospina (2016) – "Literacy". Dipublikasikan daring di OurWorldInData.org. Diambil dari: https://ourworldindata.org/literacy [Daring]
  4. ↑ Buşan AM. "Learning styles of medical students – implications in education". Curr Health Sci J. 2014;40(2):104-110. doi:10.12865/CHSJ.40.02.04
Retrieved from "https://meta.wikimedia.org/w/index.php?title=Wikispeech/id&oldid=31159657"