BIAS TC participants –
I’ve been contacted by the organizers of a W3C workshop
on speaker identification & verification web services to determine our
interest in participating/collaborating. I’m attaching the workshop
information below. This will be a discussion topic for our next meeting on 13
November. Please feel free to post any thoughts you may have on it beforehand
as well.
Regards,
CT
<http://www.w3.org/>
Workshop on Speaker biometrics and VoiceXML 3.0
5-6 March 2009
Hosted by SRI International,
Sponsors
SRI International
Background
The W3C Voice Browser Working Group <http://www.w3.org/Voice/>
seeks to develop standards to support secure access to the Web and Web
services using biometric, speaker identification and verification (SIV).
Interest in SIV is growing in both the private and public sector. That interest
is motivated by a variety of factors, primarily cost and labor issues;
convenience; and the growing number of regulations/laws governing data privacy
and security that have been put in place exist at international, national,
local, and industry levels.
Unlike other biometric technologies, speech recognition, and
speech synthesis, there are no standards specifically governing the use of SIV.
ISO/IEC 19784-1 (called “BioAPI”) is a generic, biometric
application programming language that was designed to support SIV in
non-telephony deployments. Its utility for SIV Web-services applications has
not yet been fully explored. The three other SIV standards projects (Media
Resources Control Protocol (MRCP V2) - Internet Engineering Technology Forum;
NCITS 1821-D Speaker Recognition Format for Raw Data Interchange - VoiceXML
Forum & InterNational Committee for Information Technology Standards; and
ISO/IEC 1.37.19794-13, Voice Data - International Standards Organization and
the International Electrotechnical Commission) are all still under development.
Despite the growing interest in SIV, the lack of standards
– especially for applications using Web Services – is a market and
technology barrier.
Workshop Goals
The goal of this workshop is to identify and prioritize
directions for SIV standards work as a means of making SIV more useful in
current and emerging markets. The Voice Browser Working Group is considering
following three activities:
- Identify application requirements for SIV in
VoiceXML 3.0
- Identifying SIV standards relevant to VoiceXML
3.0
- Integrate existing and in-process standards with
VoiceXML 3.0
The main outcome of the workshop will be the publication of
a document that will serve as a guide for meeting thee above goals.
Scope of the Workshop
This
workshop is focused on SIV within VoiceXML 3.0. The scope of this workshop is
restricted in order to make the best use of participants' time. In general,
discussion at the workshop and in the position papers should stay focused on
the workshop goal. We may identify new standards for SIV in addition to the
existing or in-process SIV standards.
We expect SIV to
support a wide variety of privacy policies, but the topics of the relative merits
of various privacy policies and which policies are subject to legislation are
outside the scope of this workshop.
Expected Participants
We expect
several communities to contribute to the workshop:
- SIV technology vendors
- Developers of applications using SIV
- Biometric specialists interested in
incorporating SIV into their systems
- Security industry specialists
- SIV researchers and other experts
- Users of SIV in the public and private sectors
- Standards bodies interested in SIV
Requirements
for Participation
Position
papers will be the basis for the discussions at the workshop. Each organization
or individual wishing to participate must submit a position paper by the date
shown below.
All workshop
attendees must submit a position paper. Papers
must be submitted by email to member-siv-submit@w3.org
before 18 December 2008.
Papers should:
- Explain the participant's interest in the
workshop
- Explain their view point
- Include concrete examples of their suggestions
Position papers should be written in English. Examples may
be illustrated with non-English languages with an English explanation. All
papers should be 1 to 5 pages, although they may link to longer versions or
appendices. Allowed formats are valid HTML or XHTML, PDF, or plain text. Papers
in any other format (including invalid HTML/XHTML) will be returned with a
request for correct formatting.
Participation in
the workshop is conditional upon acceptance of the position paper by the
program committee.
Accepted
position papers will be published on the public Web page of the workshop.
Submitting a position paper comprises a default recognition of these terms for
publication.
The Program
Committee will ask the authors of particularly salient position papers to
explicitly present their position at the workshop to foster discussion.
Presenters will be asked to make the slides of the presentation available on
the workshop home page in HTML, PDF, or plain text.
See the schedule
below for submission and registration deadlines.
Topics:
Possible
topics include, but are not limited to the following:
- Integration of XML and nonXML formats
- Should speaker biometric data
be represented in binary or xml formats? Should both formats be
standardized?
- Should developers use a
programming API or an XML-based API? Should both formats be standardized?
- Can commands, events, and data
structures be normalized or standardized?
- What is the relationship with
MRCP V2 and how should that relationship evolve?
- What is the relationship with
BioAPI and how should that relationship evolve?
- Can audio formats be
normalized for interoperability and conformance testing? How to determine
that normalization?
- Security data
- What mechanisms should be used
to maintain the security of voice models and voice model databases?
- What mechanisms should be used
to maintain the additional nonbiometric security data?
- What mechanisms should be used
to secure the transferring of voice biometric data between client and
server?
- The role of multimodal biometrics
- How to combine biometric
and/or nonbiometric techniques to reach authentication decisions?
- What is the role of multimodal
technology in authentication?
- What is the role of EMMA or
other multimodal annotation techniques?
- Other topics relevant to authentication
This workshop is focused on SIV within VoiceXML 3.0
Participation:
Participation
will be governed by the following:
- To ensure maximum diversity, the number of
participants per organization will be limited.
- W3C membership is not required to participate in
this workshop.
- Workshop sessions and documents will be in
English.
- Attendees are required to submit a position
paper.
Workshop
Organization
Workshop Chairs:
- Judith Markowitz <mailto:judith@jmarkowitz.com> ,
co-Chair Speaker Biometrics Committee (VoiceXML Forum),
- Ken Rehor <mailto:ken@rehor.com>
, co-Chair Speaker Biometrics Committee (VoiceXML Forum),
- Kazuyuki Ashimura <mailto:ashimura@w3.org> , Voice
Browser Activity Lead (W3C),
Program Committee:
The current
program committee consists of:
- Cathy Tilton, Daon Corporation
- Patrick Grother, National Institute of Standards
in Technology (NIST)
- Jim Wayman, San Jose State University/Co-Chair
ISO SC 37 Special Group on Speaker Recognition
- Saurav Bhattacharyya, Viometrix Private
Unlimited/Chair ISO SC 37 Special Group on Speaker Recognition
- James A. Larson, co-chair, W3C Voice Browser
Working Group
- Dave Raggett, W3C/Justsystems
- Daniel C. Burnett, Voxeo
- Tomoko Matsui, Institute of Statistical
- Alvin Martin, Speech Group of the National
Institute of Standards in Technology (NIST)
Schedule:
The workshop
program will run from 8:30 am to 6 pm on both days.
Venue:
The workshop
will be held at SRI International,
Information
about hotels will be distributed with acceptance letters.
Registration:
Information
on registration, hotel and logistics will be sent with the notification of
acceptance.
Important Dates
Date Event
18
December 2008 Deadline for position papers. Submit position papers
to member-sivws-submit@w3.org
<mailto:member-sivws-submit@w3.org>
.
2
February 2009 Acceptance notification and registration instructions
sent. Program and accepted position papers posted on the workshop website.
16
February 2009 Deadline for registration.
5
March 2009 Workshop Begins (8:30 AM)
6
March 2009 Workshop Ends (6:00 PM)
23
March 2009 Conference minutes and conference deliverables posted
on the workshop website.
Judith Markowitz
<mailto:judith@jmarkowitz.com>
, Ken Rehor <mailto:ken@rehor.com>
and Kazuyuki Ashimura <mailto:kazuyuki@w3.org>
, Workshop co-Chairs