The journal Signal Processing published by Elsevier is issuing a call for a special issue on latent variable models and source separation. Papers dealing with multi-talker ASR and noise-robust ASR using source separation techniques are highly welcome.



                         SIGNAL PROCESSING
               http://www.elsevier.com/locate/sigpro

                          Special issue on
           LATENT VARIABLE ANALYSIS AND SIGNAL SEPARATION

                     DEADLINE: JANUARY 15, 2011


While independent component analysis and blind signal separation have become mainstream topics in signal and image processing, new approaches have emerged to solve problems involving nonlinear signal mixtures or various other types of latent variables, such as semi-blind models and matrix or tensor decompositions. All these recent topics lead to new developments and promising applications. They are the main goals of the conference LVA/ICA 2010 which took place in Saint-Malo, France, from September 27 to 30, 2010.

The aim of this special issue is to provide up to date developments on Latent Variable Analysis and Signal Separation, including theoretical analysis, algorithms and applications. Contributions are welcome both from attendees of the above conference and from authors who did not attend the conference but are active in these areas of research.

Examples of topics relevant to the special issue include:
- Non-negative matrix factorization
- Joint tensor factorization
- Latent variables
- Source separation
- Nonlinear ICA
- Noisy ICA
- BSS/ICA applications: image analysis, speech and audio data, encoding of natural scenes and sound, telecommunications, data mining, medical data processing, genomic data analysis, finance,...
- Unsolved and emerging problems: causality detection, feature selection, data mining,...

SUBMISSION INSTRUCTIONS:
Manuscript submissions shall be made through the Elsevier Editorial System (EES) at
http://ees.elsevier.com/sigpro/
Once logged in, click on “Submit New Manuscript” then select “Special Issue: LVA” in the “Choose Article Type” dropdown menu.

IMPORTANT DATES:
January 15, 2011: Manuscript submission deadline
May 15, 2011: Notification to authors
September 15, 2011: Final manuscript submission
December 15, 2011: Publication

GUEST EDITORS:
Vincent Vigneron, University of Evry – Val d’Essonne, France
Remi Gribonval, INRIA, France
Emmanuel Vincent, INRIA, France
Vicente Zarzoso, University of Nice – Sophia Antipolis, France
Terrence J. Sejnowski, Salk Institute, USA

Top

IEEE Signal Processing Magazine
Special Issue on Fundamental Technologies in Modern Speech Recognition
		 
Guest Editors:		 
Sadaoki Furui   Tokyo Institute of Technology, Tokyo, Japan  
                (furui@cs.titech.ac.jp)
Li Deng         Microsoft Research, Redmond, USA (deng@microsoft.com)
Mark Gales      University of Cambridge, Cambridge, UK (mjfg@eng.cam.ac.uk)
Hermann Ney     RWTH Aachen University, Aachen, Germany
                (ney@cs.rwth-aachen.de)
Keiichi Tokuda  Nagoya Institute of Technology, Nagoya, Japan 
                (tokuda@nitech.ac.jp)

Recently, various statistical techniques that form the basis of fundamental technologies underlying today’s automatic speech recognition (ASR) research and applications have attracted new attentions. These techniques have significantly contributed to progress in ASR, including speaker recognition, and their various applications.  The purpose of this special issue is to bring together leading experts from various disciplines to explore the impact of statistical approaches on ASR.  The special issue will provide a comprehensive overview of recent developments and open problems.

This Call for Papers invites researchers to contribute articles that have a broad appeal to the signal processing community.  Such an article could be for example a tutorial of the fundamentals or a presentation of a state-of-the-art method.  Examples of the topics that could be addressed in the article include, but are not limited to:
 * Supervised, unsupervised, and lightly supervised training/adaptation
 * Speaker-adaptive and noise-adaptive training
 * Discriminative training
 * Large-margin based methods
 * Model complexity optimization
 * Dynamic Bayesian networks for various levels of speech modeling and decoding
 * Deep belief networks and related deep learning techniques
 * Sparse coding for speech feature extraction and modeling
 * Feature parameter compensation/normalization
 * Acoustic factorization
 * Conditional random fields (CRF) for modeling and decoding
 * Acoustic source separation by PCA and ICA
 * De-reverberation
 * Rapid language adaptation for multilingual speech recognition
 * Weighted-finite-state-transducer (WFST) based decoding
 * Uncertainty decoding
 * Speaker recognition, especially text-independent speaker verification
 * Statistical framework for human-computer dialogue modeling
 * Automatic speech summarization and information extraction

Submission Procedure:
Prospective authors should submit their white papers to the web submission system at http://mc.manuscriptcentral.com/spmag-ieee.

Schedule:
 * White paper due:         October 1, 2011
 * Invitation notification: November 1, 2011
 * Manuscript due:          February 1, 2012
 * Acceptance notification: April 1, 2012
 * Final manuscript due:    May 15, 2012
 * Publication date:        September 15, 2012 

Top

 COMPUTER SPEECH AND LANGUAGE
                 http://www.elsevier.com/locate/csl

                          Special issue on
   SPEECH SEPARATION AND RECOGNITION IN MULTISOURCE ENVIRONMENTS 

                  *FINAL DEADLINE: DECEMBER 31, 2011* 


One of the chief difficulties of building distant-microphone speech recognition systems for use in `everyday' applications is that the noise background is typically `multisource'. A speech recognition system designed to operate in a family home, for example, must contend with competing noise from televisions and radios, children playing, vacuum cleaners, and outdoors noises from open windows. Despite their complexity, such environments contain structure that can be learnt and exploited using advanced source separation, machine learning and speech recognition techniques such as those presented at the 1st International Workshop on Machine Listening in Multisource Environments (CHiME 2011).


This special issue solicits papers describing advances in speech separation and recognition in multisource noise environments, including theoretical developments, algorithms or systems.


Examples of topics relevant to the special issue include:
• multiple speaker localization, beamforming and source separation,
• hearing inspired approaches to multisource processing,
• background noise tracking and modelling,
• noise-robust speech decoding,
• model combination approaches to robust speech recognition,
• datasets, toolboxes and other resources for multisource speech separation and recognition.


SUBMISSION INSTRUCTIONS:
Manuscript submissions shall be made through the Elsevier Editorial System (EES) at
http://ees.elsevier.com/csl/
Once logged in, click on “Submit New Manuscript” then select “Special Issue: Speech Separation and Recognition in Multisource Environments” in the “Choose Article Type” dropdown menu.


IMPORTANT DATES:
November 30, 2011: Paper submission
March 30, 2012: First review
May 30, 2012: Revised submission
July 30, 2012: Second review
August 30, 2012: Camera-ready submission

Top

Call for Papers

 Special Issue on Processing

Under-Resourced Languages

The creation of language and acoustic resources, for any given spoken language, is typically a costly task. For example, a large amount of time and money is required to properly create annotated speech corpora for automatic speech recognition (ASR), domain-specific text corpora for language modeling (LM), etc. The development of speech technologies (ASR, Text-to-Speech) for the already high-resourced languages (such as English, French or Mandarin, for example) is less constrained by this issue and, consequently, high-performance commercial systems are already on the market. On the other hand, for under-resourced languages, the above issue is typically the main obstacle.

Given this, the scientific community’s concern with porting, adapting, or creating language and acoustic resources or even models for low-resourced languages has been growing recently and several algorithms and methods of adaptation have been proposed and experimented with. In the mean time, workshops and special sessions have been organized on this domain.

This special issue focuses on research and development of new tools based on speech technologies for less-resourced national languages, mainly, used in the following large geographical regions: Eastern Europe, South and Southeast Asia, West Asia, North Africa, Sub-Saharan Africa, South and Central America, Oceania. The special issue is open to present problems and peculiarities of targeted languages in application to spoken language technologies, including automatic speech recognition, text-to-speech, speech-to-speech translation, spoken dialogue systems in an internationalized context. When developing speech-based technologies researchers are faced with many new problems from lack of audio databases and linguistic resources (lexicons, grammars, text collections), to inefficiency of existing methods for language and acoustical modeling, and limited infrastructure for the creation of relevant resources. They often have to deal with novel linguistic phenomena that are poorly studied or researched from a speech technology perspective (for instance, clicks in southern African languages, tone in many languages of the world, language switching in multilingual systems, rich morphology, etc).

Well-written papers on speech technologies for targeted languages are encouraged, and papers describing original results (theoretical and/or experimental) obtained for under-resourced languages, but important for well-elaborated languages too, are invited as well. Good papers from any countries and any authors may be accepted if they present new speech studies concerning the languages of interest of the special issue. Submissions from countries where issues related to under-resourced languages are a practical reality, are strongly encouraged for this special issue.

Important Dates:

Submission deadline:  1st August 2012

Notification of acceptance: 1st February 2013

Final manuscript due:  April 2013

Tentative publication date: Summer 2013

Editors

Etienne Barnard

North-West University, South Africa

Laurent Besacier

Laboratory of Informatics of Grenoble, France

Alexey Karpov

SPIIRAS, Saint-Petersburg, Russia

Tanja Schultz

University of Karlsruhe, Germany

Top

Call for Papers
Fax 41 61 306 12 34
E- Mail karger@karger.ch
www.karger.com
© 2011 S. Karger AG, Basel
0031–8388/11/0683–0117
$38.00/0
Accessible online at:
www.karger.com/pho
Phonetica 2011;68:117–119
DOI: 10.1159/000334686
Speech Production and Perception across
the Segment-Prosody Divide:
Data – Theory – Modelling
The concept of sound segments has traditionally played a central role in the
phonetic representation of words. It underlies the development of alphabetic writing
systems, of phonetic transcription and of phonemic theory. Other sound aspects,
especially pitch, but also energy, voice quality, rhythm have been conceptualized as
being superpositioned on segments in a broader frame of syllables and utterances. The
segment is associated with the short-time window of opening and closing movements
of the vocal tract, and simultaneously, with the differentiation of lexical and propositional
meaning, whereas prosodies are generally associated with long-time windows
of pitch, energy and voice quality control, and predominantly with attitudinal and
expressive utterance meaning, including the functions of attention seeking, intensity
signalling, and syntagmatic phrasing. This different substance-meaning duality
in sound segments and prosodies accounts for the current dichotomous mainstream
research paradigms of sounds and prosodies. The sound-prosody dichotomy has,
however, been repeatedly called into question, among others in the Firthian School
of Prosodic Analysis, following Firth’s seminal article ‘Sounds and Prosodies’ from
1948, for example in the study of such suprasegmental phenomena as vowel harmony
or long articulatory components of, e.g., palatalization, velarization, nasalization,
glottalization in the linguistic function of distinctively marking words and morphological
structures. Moreover, it has always been bridged in the analysis of lexical
stress, where segmental aspects of vowel duration and vowel spectrum, and prosodic
aspects of fundamental frequency and energy have jointly been taken into account in
a vast array of experimental investigations.
The reliance on linguistic form and phonetic substance in the analysis of sound
segments and prosodies reflects the tenets of 20th century structural linguistics, as it
relegates the functional aspect of speech communication to a post hoc level. Such a
dichotomous formal approach is a useful heuristics to come to grips with the enormous
complexity of speech, especially in the initial stages in the investigation of a language.
Yet, the formal manifestations, analytically separated as sounds and prosodies, are
the joint expression of the manifold communicative functions in speech: semantic,
information-structural, expressive and attitudinal. If these functions are taken as the
superordinate control variable, the axiomatic formal dichotomy of sounds and prosodies
fades away because they interact, with varying weights, in the coding of specific
communicative functions. This functional approach to phonetic detail in segmentprosody
interaction was the empirical and theoretical theme of two recent plenary
talks at the 17th ICPhS in Hong Kong: ‘Does Phonetic Detail Guide Situation-Specific
Speech Recognition?’ by Sarah Hawkins and ‘On the Interdependence of Sounds and
Prosodies in Communicative Functions’ by Klaus Kohler. They were preceded by papers
01_PHO334686.indd 117 17/11/11 22:08:26
118 Phonetica 2011;68:117–119 Call for Papers
in Journal of the Acoustical Society of America, Journal of Phonetics, and Phonetica:
Niebuhr, O.: Coding of intonational meanings beyond F0: evidence from
utterance-fi nal /t/ aspiration in German. J. acoust. Soc. Am. 124: 1252–1263
(2008).
Hawkins, S.: Roles and representations of systematic fi ne phonetic detail in
speech understanding. J. Phonet. 31: 373–405 (2003).
Local, J.: Variable domains and variable relevance: interpreting phonetic
exponents. J. Phonet. 31: 321–339 (2003).
Kohler K.: Communicative functions integrate segments in prosodies and
prosodies in segments. Phonetica 68: 26–56 (2011).
Kohler, K.; Niebuhr O.: On the role of articulatory prosodies in German message
decoding. Phonetica 68: 57–87 (2011).
On the one hand, these investigations showed systematic phonetic detail in talkin-
interaction as well as acoustic effects of segments on pitch patterns and of pitch
patterns on segments in the perceptual identification of semantic functions, and, on the
other hand, demonstrated the perceptual importance of long phonetic components of,
e.g., palatalization that are not linked to a segmentable sound unit but are superimposed
as an articulatory prosody on a wider stretch of speech.
We would like to make this sound-prosody relationship the theme of a special
issue of Phonetica and raise the central question:
How are sounds and prosodies intertwined, mutually shaping each other,
as a reflection of different communicative functions in speech interaction?
The papers we solicit are to take a renewed look in greater breadth and detail at this
interweaving of the threads of sounds and prosodies in a tapestry of speech communication
in a variety of languages, incorporating all forms of meaning – propositional,
attitudinal and expressive. The guiding principles for submissions are as follows:
• Papers present single-language or comparative analyses of new data in a variety of
languages that highlight the interdependence of short- and long-time windows of
speech production and/or perception in relation to specifi c communicative functions
or
they discuss aspects of the theory of segment-prosody interdependence based
on language-specifi c, typological or universal relations between communicative
function and phonetic substance
or
they attempt to model segment-prosody interaction in these function-substance
relations, for example in developing algorithms for contextually and situationally
adequate high-quality speech synthesis.
• Data can be either experimental or from corpora, unscripted ones in particular, and
experimentally collected items of speech need to be functionally and situationally
anchored, which rules out the widespread metalinguistic sentence frame of the
type ‘Say X again.’, commonly used in EMMA and EPG data acquisition.
• Potential topics may include:
– segment-prosody interdependence in talk-in-interaction,
– prosodic and segmental properties in the manifestation of speech functions,
for example different types of emphasis, in production and perception,
– contribution of vowel spectrum to lexical stress perception,
01_PHO334686.indd 118 17/11/11 22:08:26
Speech Production and Perception across the Phonetica 2011;68:117–119 119
Segment-Prosody Divide
– spectral shaping of segments, for example fricatives and plosive releases, in
falling or rising f0 contours, and perceptual effects,
– articulatory prosodies in speech reduction, especially of function words, and
their importance in speech decoding,
– creation of rhythmic fl ow by preferred segmental patterns, such as high
versus low vowels, rather than the reverse, in fl ip-fl op, sing song, ping-pong,
zig zag, wishy washy, or avoidance of phrase-internal obstruent breaks in
sonorant stretches, as in thunder and lightning against the semantically
obvious *lightning and thunder, or mum and dad, German Mama und Papa,
Oma und Opa,
– signalling of tone and intonation in whispered speech
– contribution of segments and prosody to the generation of high-quality
speech synthesis and to (online) spoken word recognition.
Editorial Guidelines and Schedule
The total space available will be a double issue of the Journal. We expect to publish
approximately 12 contributions of 12 printed pages each on average. Submissions
need to follow the Phonetica style sheet (cf. ‘Instructions to Authors’ in any recent
issue and www.karger.com/electronic_submission) and should include Word and pdf
files. The dates of the editing schedule are as follows:
By 28 January, 2012: Submission by e-mail attachment to kjk@ipds.uni-kiel.de of
an 800-word abstract, giving title, author(s), affiliation(s),
e-mail address of main author.
29 February, 2012: Notification of authors whether the proposed papers have been
recommended as potential contributions to the theme by the
Editorial Team, and, if so, invitation to submit full versions for
review.
By 31 May, 2012: Electronic submission of pdf files as e-mail attachments to
kjk@ipds.uni-kiel.de, to be sent out for review.
31 July, 2012: Intimation of final decision about acceptance for publication in
the special issue, including reviewers’ comments and suggestions
for revision. Due to the tight publication schedule only
papers requiring minor or moderate revision can be included
in the special issue. If major revision is necessary, authors will
be encouraged to resubmit for publication in an ordinary issue
of Phonetica.
By 20 August, 2012: Submission of final versions in Word and pdf by e-mail attachment
to kjk@ipds.uni-kiel.de.
End of 2012: Publication.

Top