Submissions/Increasing Wiki Accessibility: Role of Speech Technologies

From Wikimania 2013 • Hong Kong
Jump to navigation Jump to search
Xmark01.svg

After careful consideration, the programme committee has decided not to accept the below submission at this time. Thank you to the author(s) for participating in the Wikimania 2013 programme submission, we hope to still see you at Wikimania this August.

Submission no.
5015
Subject no.
B3


Title of the submission
Increasing Wiki Accessibility: Role of Speech Technologies
Type of submission
Presentation
Author of the submission
Basil George
Country of origin
India
Affiliation
Speech and Vision Lab, IIIT- Hyderabad
E-mail address
basilgeorge007@gmail.com
Username
BasilGeorge22
Personal homepage or blog
Personal homepage, English blog, Malayalam blog
Abstract

Increasing Reach and Increasing Participation are top Wikimedia strategies. Though amazing personalities like Aniruddha Kumar actively contributes to Wikimedia unperturbed by his disability, a large segment of users still find the current state of the art open-source assistive technologies lacking in improving their access to free knowledge resources. India has huge visually challenged population. When this is coupled with high levels of illiteracy, a large demographic gets alienated to the free access of knowledge. In this presentation, I try to share my ideas, as a speech processing researcher, on how to increase the accessibility of Wiki using various speech technologies such as audio search, speech recognition and speech synthesis, focusing on Indian languages in particular.

Detailed proposal

Integration of speech technologies has a huge potential to increase the accessibility of Wikipedia to new users and contributors. Currently, visually challenged (VC) users are greatly disadvantaged when it comes to participating in the free knowledge movement chiefly due to the lack of assistive technologies (AT). Studies suggest that about 285 million are VC worldwide with India being the home to a majority of them. Another challenge in the Indian context is the high illiteracy rate. Hence, we need assistive technologies which help them be part of the free knowledge network.

Lot of R&D is happening in the areas of speech recognition and speech synthesis. Currently, some open source screen readers such as Orca and NVDA are available which reads out what is displayed on the screen. But the lack of quality of the sounds produced and their lack of support in most of the languages of the world make their use limited. Similar is the case with speech recognition projects such as CMU Sphinx and VoxForge. For making Wiki fully accessible, voiced-based search capabilities coupled with speech recognition and synthesis needs to be integrated into the Wiki framework. The progress made towards achieving this goal in the Indian context will be discussed in this presentation.

The challenges are not few. This gives a lot of room for development and a huge scope for community participation. WikiProject Spoken Wikipedia is a huge attempt in this direction. Such efforts need to be initialized for Indian languages too. Manual recording of all articles in Wikipedia may not be possible, but such an effort will not only increase the reach, but also serves to create an excellent speech corpus for speech technology research. Computer generated voice quality can be significantly improved by building better pronunciation dictionaries with community participation. Inclusion of voice search capabilities in Spoken Wikipedia would revolutionize the way people experiences Wikipedia.

Track
Technology and Infrastructure, Wikis in Asia
Length of presentation/talk
25 Minutes
Language of presentation/talk
English
Will you attend Wikimania if your submission is not accepted?
Yes, depending on the scholarship.
Slides or further information (optional)
To be uploaded.
Special requests
None


Interested attendees

If you are interested in attending this session, please sign with your username below. This will help reviewers to decide which sessions are of high interest. Sign with four tildes. (~~~~).

  1. Danny B.
  2. --Netha Hussain (talk) 09:24, 18 February 2013 (UTC)
  3. See also Submissions/Making wiki accessible for visually impaired editor. Blue Rasberry (talk) 08:38, 19 February 2013 (UTC)
  4. As someone interested in making some spoken articles for the English Wikipedia, this interests me. CT Cooper · talk 23:49, 28 April 2013 (UTC)
  5. John Andersson (WMSE) (talk) 15:23, 3 May 2013 (UTC) I am very interested in this topic.
  6. As a blind Wikipedia editor, I think this is an interesting take on the issue. Graham87 (talk) 12:44, 5 May 2013 (UTC)
  7. Ijon (talk) 23:50, 15 May 2013 (UTC)
  8. Add your username here.