Wikispeech/Wikispeech 2016


Wikispeech 2016, the first Wikispeech project, aimed to create an open source TTS (text-to-speech) tool, to make the Wikimedia projects more accessible. This would help a lot of people who, for different reasons, prefer to access information by listening rather than reading. The focus was on Wikipedia, being the most well used among them and containing primarily written information. That said, there are plenty of other applications that could benefit from a freely available TTS solution, both among and outside the Wikimedia projects.

Wikispeech 2016 went from March 2016 to September 2017, preceded by a pilot study in 2015, and was financed by the Swedish Post and Telecom Authority (PTS). During this time, a MediaWiki extension was developed alongside a TTS service called Speechoid.

At the end of the project, it had support for English, Swedish and Arabic and was built modularly to allow adding more languages and voices later on. The interface allows the user to listen to an article, navigate by skipping back and forth and includes highlighting the text that is being read. A demo wiki was set up running the development versions of the extension and Speechoid.[1]

We will continue the development in Wikispeech 2019 with the goal to activate Wikispeech Text-to-Speech on the relevant language versions of Wikipedia.

The implementation of Wikispeech as of the end of Wikispeech 2016.

About

edit

With the help of synthetic speech, people who find it easier to assimilate information through speech than text can get equal access to the information. In the long run the open nature of the project will make it possible to develop new ways of presenting spoken information, e.g. through a player intended for mobile phones. This could include those with visual impairment, dyslexia or who are not literate. The approximately 25% of people who find it easier to learn from spoken text could utilize this functionality as well as those who wish to learn at the same time as they do something else (e.g. driving). 25% of the readers of Wikipedia would mean that approximately 115–125 million people would benefit from the project in the long run.

Those who have received a medical diagnosis regarding limitations in reading comprehension (e.g. dyslexia, visual impairment or cognitive impairment) often have access to technological aids. This however very often requires a diagnosis, that you live in a high income country and that the language you speak has working text-to-speech for this to be a solution to the accessibility problem. People with poor reading comprehension (from unaccustomed readers to analphabets) also have limited access to commercial tools even if this would improve their understanding. This is especially true if they do not wish to share their data with one of the IT behemoths. To conclude, the assessment is that a very large group would benefit from built in text-to-speech on Wikipedia. Making all of the websites which use MediaWiki more accessible to those who find it hard to assimilate written information is therefore incredibly important.

The project will increase the accessibility of one of the most important websites. All other platforms using MediaWiki will be able to make use of the technical solutions which are developed during the project. That is several thousand websites which quick and easy will be able to activate text-to-speech.

See Also

edit

Wikispeech/Wikispeech 2016/FAQ

I like this project...

edit

... and I would like to help out

edit

Please add yourself below if you are interested in helping out with the project. Please specify who you are and your areas of interest and any relevant expertise.

  • YOUR NAME − WHAT YOU CAN HELP WITH
  • Heath Ahrens (hahrens at ispeech dot org)- can help with text to speech in 27 languages. Founder of http://www.iSpeech.org
  • brion (talk) - browser side stuff and mediawiki integration :)
  • Pawel Cyrta – ex-Samsung s-voice tts engineer for european languages
  • Andrew Krizhanovsky – can test application in Russian language with my students
  • High-five for bringing such an amazing project. Would love to contribute for the Odia language. --Psubhashish (talk) 11:07, 22 December 2016 (UTC)[reply]
  • Bodhisattwa - Interested to have this feature for Bengali language wikis.
  • Arabic --Zack (talk) 15:26, 2 January 2017 (UTC)[reply]
  • Dr Mel Ganus - working with American Foundation for the Blind - community building and engagement --DrMel (talk) 19:53, 17 January 2017 (UTC)[reply]
  • Alangi Derick (alangiderick at gmail dot com) - I would like to help out as a volunteer developer on the project. --Alangi derick (talk) 12:09, 24 February 2017 (UTC)[reply]
  • I love this project and will contribute in Telugu language. --Pavan santhosh.s (talk) 06:28, 20 March 2017 (UTC)[reply]
  • Egbe Eugene (agboreugene at gmail dot com) - This is an interesting project and I will like to help as a volunteer developer in building tools for other volunteers to make it easier to improve the Wikispeech lexicon --Egbe Eugene (talk)
  • Robin Owain, Wikimedia UK Manager in Wales; Cymraeg / Welsh language; Admin/Bureaucrat on cywiki; browser side up. Llywelyn2000 (talk) 23:02, 17 May 2017 (UTC)[reply]
  • SEAN patrick- aspiring linguist/teacher. Can at least understand Espanol and speak English. Only 21 years young and interested in language in all sorts of ways from books to people all over the world. sean.creamer@comcast.net
  • I may not know the answers but it's 2017 I know where to find them. Shawshank
  • Moritz Schubotz - Help with support for math. --Physikerwelt (talk) 09:18, 31 May 2018 (UTC)[reply]
  • Floblack - Can help with italian language and user testing

... and I would like to endorse it

edit

Please add your name and rationale for endorsing this project in the list below. Other feedback, questions or concerns from community members are also highly valued, but please post them on the talk page.

  • Support, Recording Audio is time consuming and developing a TTS would improve accessibility. Will there be considerations made for other projects with specalist markup like Wikisource? (A lot of formatting in templates for example).

I'll also note that IIRC Edinburgh Universaity had done a LOT of work of so called voice banking, so they may be a useful academic partner to approach. ShakespeareFan00 (talk) 13:10, 9 February 2017 (UTC)[reply]

Footnotes

edit
  1. Note that this demo is still being updated as part of the 2019 project.