The Wikispeech pilot project is a joint effort between the KTH Royal Institute of Technology, Wikimedia Sweden and STTS speech technology services, with further assistance and financial backing provided by the Swedish Post and Telecom Authority.
Joakim Gustafson, professor of Speech Technology at KTH, notes that the goal of the project is to provide access to Wikipedia content for people who are blind or visually impaired, as well as people with reading difficulties. Wikimedia Sweden estimates that 25 percent of all Wikipedia users (nearly 125 million people per month) require or prefer text in spoken form.
Wikipedia’s existing crowdsourcing format, which allows users to freely contribute content under the rules of Wikimedia Commons, will be the basis of providing additional language support for the platform.
"Initially our focus will be on Swedish language, where we will make use of our own language resources," says Gustafson. "Then we will do a basic English voice, which we expect to be quite good given the large amount of open source linguistic resources. And finally, we will do a rudimentary Arabic voice, that will be more a proof of concept."
The English, Swedish and Arabic speech engines are estimated to be available by September 2017, with the remaining 280 Wikimedia-supported languages to be provided over time with community assistance.
Further information is available via the KTH and Wikipedia develop first crowdsourced speech engine article on the KTH website.
Top of page