Content uploaded by Langis Gagnon
Author content
All content in this area was uploaded by Langis Gagnon
Content may be subject to copyright.
Accessible Videodescription On-Demand
Claude Chapdelaine and Langis Gagnon
R&D Department, Computer Research Institute of Montreal (CRIM)
550 Sherbrooke St.West, Suite 100
Montreal (Quebec), H3A1B9, CANADA
(514) 840-1234
{claude.chapdelaine; langis.gagnon}@crim.ca
ABSTRACT
Providing blind and visually impaired people with the
descriptions of key visual elements can greatly improve the
accessibility of video, film and television. This project presents a
Website platform for rendering videodescription (VD) using an
adapted player. Our goal is to test the usability of an accessible
player that provides end-users with various levels of VD, on-
demand. This paper summarizes the user evaluations covering 1)
the usability of the player and its controls, and 2) the quality and
quantity of the VD selected. The complete results of these
evaluations, including the accessibility of the Website, will be
presented in the poster. Final results show that 90% of the
participants agreed on the relevancy of a multi-level VD player.
All of them rated the player easy to use. Some improvements were
also identified. We found that there is a great need to provide
blind and visually impaired people with more flexible tool to
access rich media content.
Categories and Subject Descriptors
H.5.2 [User Interfaces]: Evaluation/methodology; K.4.2 [Social
Issues]: Assistive technologies for persons with disabilities.
Keywords
Rich media, Web accessibility, audio description, blind and visual
impairment.
1. INTRODUCTION
For the blind and visually impaired, the enjoyment of visual
media can be largely improved by the addition of narrative
descriptions corresponding to the relevant visual element.
Videodescription (VD), also known as audio description or
described video, is delivered through an audio channel that
enables the blind and visually impaired to form a more accurate
and vivid mental representation of what is shown on the screen.
However, essential questions such as: what are the key elements
to be convey, how can they be best described and how many can
there be in the gaps between existing speech segments, are not
fully answered when producing VD. Research on VD issues
[1][2][3] and guidelines on production practices [4][5] are
emerging. Yet, more research is needed in order for VD to be
known and used as much as captioning is for the deaf and hearing
impaired.
Copyright is held by the author/owner(s).
ASSETS’09, October 25–28, 2009, Pittsburgh, Pennsylvania, USA.
ACM 978-1-60558-558-1/09/10.
This paper presents the accessibility analysis of a Website
platform for rendering videodescription (VD) using an adapted
player, called VDPlayer. It is an initiative to promote and improve
the richness of the multimedia experience for people with vision
impairments.
2. ACCESSIBLE VIDEO
The challenges of producing accessible video for the blind and
visually impaired are numerous. Human issues relate to the
usability of what should be described and how much is needed.
While the accessibility issues are closely coupled with the
rendering medium; we will focus mainly on the Web environment.
Human Issues. The blind and the visually impaired do not
appraise their needs on VD at the same level. For example, some
mentioned that in real life situation when no VD is available, they
stop the DVD and ask questions to their sighted friends. For them,
the perfect rendering of VD would be to have the same freedom
without the constant need of a sighted friend. In recent user’s
consultation works [2], video with varying quantity and quality of
VD were shown to 30 participants. The resulting discussions
revealed that participants had preferences depending on his/her
level of blindness, personal taste and experience. A challenge
arises when a user desires more VD that the one delivered in the
available time. Actually, the Web Content Accessibility
Guidelines (WCAG) [6] proposes to offer extended VD when
foreground audio is insufficient. But, how to implement an
extended version of VD became one of the specifications of our
project [7].
Accessibility Issues. Implementing an accessible Website without
any rich media is a laborious task in itself, since each browser
implemented the W3C recommendations [6] with its own flavor.
It becomes a greater challenge when dealing with rich media
objects even with the available resources [8]. Furthermore,
problems are still too often part of the visually impaired users’
browsing experiences [9][10][11][12]. It is a known fact that their
interactions take much longer time than sighted users [13][14]; it
can take up to three minutes to get to the main content of page
with a screen reader. Solutions involved developing tool to
measure a Website’s conformance to the guidelines [15]. Many
tools are available, see [16] for a compare study.
3. USERS EVALUATION
We developed an accessible Website that presented five short
films that can be screened with our VDPlayer. We produced the
VD descriptions with the rendering of a synthetic voice and
provided two levels of VD: 1) the standard mode that gave the
VD fitting in the available non-speech segment and 2) the
extended mode that offered all VD produced which exceeded the
non-speech segments. In this later case, it required the player to
stop, deliver the VD and restart. The VDPlayer was designed to
1) offer standard video controls such as play, pause, rewind,
forward and volume change, and 2) provide controls specific to
the VD. Such as, to allow users to select the level of VD they
wanted, to personalize the item of VD that they could hear or to
repeat the last item of VD that was said.
After each viewing, participants could fill up an evaluation
questionnaire to provide feedback on the VD, their interaction
with the player and the Website. Ten participants completed the
questionnaire. The group was composed of seven men and three
women and most of them were over 31 years old. All the
participants navigated with a screen reader (9 with Jaws and 1
with Window Eye). All of them considered themselves an expert
user of their tool. Most participants declared being frequent
television viewers but infrequent users of VD mainly imputable to
its low availability.
3.1 VDPlayer Evaluation
The VDPlayer evaluation aimed at establishing the perceived
relevancy and ease of use of the video player controls related to
VD.
Relevancy. Selecting various VD levels was found relevant to
some degree (strongly and fairly) by 90% of the participants.
None of them found it to be irrelevant. Many participants
mentioned that they listened to more than one version and
appreciated more the extended version. This high score suggests
there is a need for this type of functionality.
Ease of use was evaluated for the player in general and for each of
the VD controls. In general, the player was judged fairly easy to
use by 90% of the participants. Individual controls were rated ease
to use by 90% to 100% of the participants. One participant who
rated the player fairly difficult to use commented that the sound
level was very low and that he was unable to augment it. After
verification, we found the related technical problem and corrected
it.
3.2 VD Evaluation
The evaluation of VD itself was done through a series of nine
statements either having a positive or negative tone. Participants
had to choose the level of agreement or disagreement for each of
them. The produced VD was greatly appreciated by the
participants. Indeed, 92% agreed (strongly or fairly agreed) with
the positive statements. While, only 15% agreed with the negative
statements. The weakest scores were related to 1) the quality of
the synthetic voice for which three participants judged it to be
moderately unacceptable, 2) most participants had the impression
the VD covered relevant audio information and one participant
found the VD frustrating at times. In conclusion, global results
indicate that VD is good and corresponds to a need but some
improvements could be implemented to better convey VD to the
listeners.
4. DISCUSSIONS
The goals of our project were reached since video with its
different levels of VD were made available and were screened by
blind or visually impaired users. Further, the user evaluation
proved that our approach was accessible and corresponded to their
needs. More recently, we have integrated a keyboard logging
algorithm into our VDplayer to better analyses how navigating in
the player is done by blind and visually impaired users in order to
better measure ease of use. In the future, more user testing is
planned to evaluate the robustness and usability of an interactive
accessible VDPlayer.
5. REFERENCES
[1] Piety, P.J. 2004. The language system of audio description:
an investigation as a discursive process. JVIB 98, no 8: 1-36.
[2] Turner, J.M. and Colinet, E. 2004. Using audio description
for indexing moving images. Knowledge organization 31, no
4: 222-230.
[3] Salway, A. 2007. A corpus-based analysis of audio
description. In Media for all. Edited by Cintas, J.D., Orero P.
and Remael A. Approaches to Translation Studies, New
York, NY. 151-174.
[4] Henry, S.L. 2006. Introduction to Web accessibility.
www.w3.org/WAI/intro/accessibility.php.
[5] Office of Communication. 2000. ITC Guidance On
Standards for Audio Description: www.ofcom.org.uk/
[6] Web Content Accessibility Guidelines (WCAG). 2008.
http://www.w3.org/TR/WCAG20/
[7] Gagnon, L. & als. 2009. Towards computer-vision software
tools to increase production and accessibility of video
description for people with vision loss. UAIS.(published
online 5 February). Springer Verlag.
[8] Flash and accessibility, http://www.usability.com.au
[9] Petrie, H. and Kheir, O. 2007. The relationship between
accessibility and usability of websites. In Proc. CHI, pages
397-406, San Jose, CA, ACM.
[10] Takagi, H., Saito, S., Fukuda, K. and Asakawa, C. 2007.
Analysis of navigability of web applications for improving
blind usability. TOCHI.
[11] Miyashita, H., Sato, D., Takagi, H. and Asakawa, C. 2007.
Making multimedia accessible for screen reader users. In
proceedings of W4A’07, ACM, pp. 126-127.
[12] Smillie, D. 2005. Instant Accessibility: does it work?, RNIB,
Web Access Center, http://www.rnib.org.uk/
[13] Bigham, J., Cavender, A.C., Brudvik, J.T., Wobbrock, J.O.
and Ladner, R.E. 2007. WebinSitu: A comparative Analysis
of Blind and Sighted Browsing Behavior, In ASSETS 2007.
[14] Takagi, H., Asakawa, C., Fukuda, K. and Maeda, J. 2004.
Accessibility designer: visualizing usability for the blind. In
ASSETS ’04.
[15] W3C/WAI. 2008. Conformance evaluation of web sites for
accessibilities. www.w3.org/WAI/eval/conformance.html.
[16] Brajnik, G. 2008. A comparative Test of Web Accessibility
Evaluation Methods, p 113. In ASSETS 2008.
