Summer school

In connection to ICVPB2008, an international summer school for Ph.D. students and post doc researchers on Advanced Voice Function Assessment will be arranged August 4th-6th 2008. See also Cost meetings .

The summer school is funded by Cost 2103 Action (http://www.cost2103.eu/).

Call for Participation into a Multi-disciplinary Summer Training School on Voice Assessment 

Time: August 4-6, 2008 

Place: University of Tampere (Main Campus, LINNA Building, Lecture room K110, and a computer room K114), Tampere, Finland 

Organizer: Department of Speech Communication and Voice Research, University of Tampere, supported by EU Cost 2103 Action (Advanced Voice Function Assessment), see http://www.cost2103.eu/ 

To whom: Ph.D. students and post doc researchers in the multi-disciplinary field of voice research (with a wide spectrum of educational background like engineering, physics, mathematics, speech science, medicine, logopedics etc.) 

What: The school will concern the following themes: Anatomy and physiology of the larynx, tissue properties of the vocal folds and biomechanics of voice production, voice and speech acoustics, various research methods that can be applied in voice assessment, analysis of the validity of methods in voice assessment, and principles and methodological concerns in database acquisition. In addition to theoretical introduction in the principles of various methods like kymography, electroglottography, mathematical modelling, and acoustic analysis methods including e.g. inverse filtering, the participants will be provided with demonstrations and possibility to apply the methods in practice. The instruction relates to and presents outcomes of the international research collaboration carried out in Cost Action 2103 working groups. 

Teachers: All teachers are internationally renowned voice researchers representing different fields. Ingo R. Titze, Ph.D (Physics) from the USA, one of the leading voice scientists in the world, will be the external expert lecturer. Other teachers are management committee and working group members in Cost 2103 Action: Philippe Dejonckere, MD, Ph.D (Otolaryngology and Phoniatrics), from Belgium is the chairman of the Cost 2103 Action, Yannis Stylianou, Ph.D (Computer Science) from Greece is the vice-chairman of the action; other teachers are: Malte Kob, Ph.D.(Electrical Engineering) from Germany, Peter Murphy, Ph.D. (Electrical Engineering and Computer Science) from Ireland, Jan Svec, Ph.D (Physics) from the Czech Republic, Jean Schoentgen, Ph.D. (Physics) from Belgium, Mette Pedersen, MD, PhD. (Otolaryngology and Phoniatrics) from Denmark, Juan Ignacio Godino Llorente, Ph.D. (Electrical Engineering) from Spain and Paavo Alku, Ph:D. (Electrical Engineering) from Finland. 

How to apply: Those who wish to participate in the summer school are requested to send a two page free application in e-mail to the address: COST-SUMMERSCHOOL@uta.fiThe application should include a CV and a research plan of the applicant’s Ph.D. or post doc research and full contact information (address, telephone, e-mail address). The deadline for submission of applications is the 16th of June. 

Notification of acceptance: 25 participants will be chosen based on an evaluation of the applications, performed by an international evaluation committee. Notification of acceptance will be sent in e-mail by the 25th of June. 

Arrangements:The participation in the school is free of charge for the 25 participants chosen; coffee and lunch during the school days and accommodation (Omena Hotel, Tampere, see www.omena.com/frontpage ) will be included. The participants will be refunded their travel expenses up to 200 euros after the school. For this purpose, the participants will need to send their original tickets and receipts of them to the address: Mrs Sinikka Pyykönen, Department of Speech Communication and Voice Research, 33014 University of Tampere, Finland.

Poster presentation: The 25 students chosen will be asked to prepare a poster to present their research during the Summer School.  The maximum size of poster is 88 cm (width), 118 cm (height).

More information will be found later on the Cost 2103 website, and on the website of the 6th International Conference on Voice Physiology and Biomechanics ICVPB2008; see www.uta.fi/conference/ICVPB2008

ABSTRACTS OF COST 2103 SUMMER TRAINING SCHOOL LECTURES

Anatomy and Tissue Properties of the Larynx

Philippe Dejonckere

The macro- and micro-anatomy of the larynx will be reviewed from a particular point of view : the relation between the structure and the different laryngeal functions, particularly the phonatory (the vocal fold as an oscillator) and sphincteric functions, and with emphasis on details of tissue properties that might be important for modelists.

Several specific topics will further be addressed :

• The gender differences
• The effects of ageing on laryngeal anatomy and histology
• The neuro- and histoplasticity of the larynx, as a pathophysiological reaction to various kinds of mechanical stress and injury, as observed in overuse and /or misuse, or in exposure to smoke and some chemical agents, or to inhalation steroids
• The effects of lubrication and hydration on vocal fold tissue properties
• The tissue differentiation and apoptosis mechanisms during embryological development, as far as they can induce or influence voice changes in adult age, within the frame of the so-called "minimal congenital laryngeal abnormalities"
• The tissue transplantation (particularly autologous fat within the vocal fold) and its effects on the vocal fold vibrations within the scope of phonosurgery, as well as recent developments in knowledge about vocal fold scarring and scarring repair
  

Biomechanics of vocal fold vibration
Ingo R. Titze
        - Fundamentals of tissue mechanics
        - Muscle mechanics
        - Vocal fold posturing
        - Flow-induced self-sustained oscillation in airways

High speed videolaryngoscopy and videokymography

Jan G. Švec
The workshop will provide introduction to laryngeal imaging, high-speed
videolaryngoscopy and videokymography.  Principles and technical description of the
methods will be provided and examples of findings will be shown. Advantages,
disadvantages and perspectives of the methods will be discussed. Categorization of
laryngeal findings based on vibration features will be shown and their evaluation will
be demonstrated on various cases of voice disorders.

Electroglottogram - introduction, measurement and application

Peter Murphy

The electroglottogram (EGG) provides a measure of laryngeal conductance from which vocal fold contact area can be inferred. A brief introduction to the basis of EGG analysis and its use in speech science is presented. Following this, new measures derived from EGG signals are indicated (e.g. a measure taken from the second derivative of the EGG signal). EGG based measures are used to investigate (i) emotional voice, (ii) vocal loading, (iii) phonation type (breathy, normal, pressed) and (iv) voice level (soft, normal, loud).

Electroglottogram - extended concepts and further applications

Malte Kob

Apart from the basic applications of the EGG as device for voice characterisation an extension of the method also allows the assessment of larynx position using multiple path measurements through the neck. The concept of such multi-channel EGG devices is presented and the application of a prototype is demonstrated for singing voice analysis, glottal closure detection and visualisation of swallowing manoeuvres.

Acoustics of the human voice

Ingo R. Titze
        - Sound generation and propagation in airways
        - Source-filter theory of vowels
        - Source-filter interactions
        - Sound radiation from airway openings

Inverse filtering

Paavo Alku

Estimation and parameterisation of the glottal flow from the speech pressure signals is
discussed. Analysis of voice production with inverse filtering is demonstrated with an
easy-to-use platform, the Aparat system. Students will conduct experiments of their own
by using the Aparat software.

Functional assessment of voice : Acoustic methods I.

Jean Schoentgen, L.I.S.T., CP 165/51, Faculty of Applied Sciences, Université Libre de

Bruxelles, B-1050, Brussels, Belgium

The objective of the presentation is to put the task of speech and voice analysis into its clinical context and give an overview of existing categories of speech cues that have been examined with a view to characterising a patient's voice and speech. Topics that are discussed are stages involved in functional assessment of voice and
speech: vocal performances a clinician may wish to examine; speech tasks; nonverbal speaker data; speech material; non-invasive sensors / transducers; signal processing / heuristics; acoustic features of voice disorders with a focus on temporal cues of vocal dysperiodicity; discussion of the robustness, reliability and validity of
existing cues.

Acoustic Analysis of Voice

Peter Murphy

The acoustic voice signal is used to categorize and quantify voice quality. In this presentation the physical and perceptual relevance of acoustic indices are considered. The spectral correlates of basic phonation parameters such as open quotient, glottal pulse skewness and shimmer are investigated using a graphical Fourier analysis representation. Spectral and cepstral indicators of voice quality are detailed and future directions are considered.

Issues in the synthesis of disordered voices

J. Schoentgen, S. Ben Elhadj Fraj, F. Grenez

Laboratory of Image & Signal Processing and Telecommunication Devices, Faculty of Applied Sciences, Université Libre de Bruxelles & National Fund for Scientific Research, Belgium

The objective of the lecture is to examine issues in the synthesis of disordered voices, which designates the numerical simulation of pathologic voice timbres. The simulation of non-modal voice may contribute to the discovery of acoustic cues of the perceptual traits of raucity, breathiness, strain, etc., as well as facilitate the training of clinicians evaluating disordered voices perceptually. The synthesizer of disordered speech, which is presented has been used to simulate a wide range of normal and disordered vocal timbres of sustained speech sounds, including (ab)normal jitter, shimmer, tremor and spasmodic dysphonia, breathiness, diplophonia and irregular phonation. Results of perceptual experiments with synthetic stimuli are also discussed.

Evidence based medical training aspects, the validity of various voice parameters.

Mette Pedersen, http://www.mpedersen.org/

Evidence in voice research is a well known aspect based on prospective randomized controlled trials (RCT). What is new is, that once you have more than one RCT, you can get even better evidence in a statistical meta analysis. This is done in cancer and cardiac research, and seems to be necessary in our field.

We have made RCTs on fundamental frequency of sustained tones, and during reading of a standard text in Danish. Jitter and shimmer were also analysed. With the electroglottography, the Qx, the laryngeal jitter and shimmer were also studied in an RCT. The problem here is that our studies are made in Danish so we have to discuss evidence of voice research with the whole European Union in the future.

References: all on website mpedersen.org

Pedersen 2000

Pedersen M, McGlashan J (2000, protocol, 2001 review). Surgical versus non-surgical interventions for vocal cord nodules. The Cochrane library. Oxford. Wiley publishing. Update 2007.

Pedersen 2004

Pedersen M . Interactions between basic and clinical research. Int. Conf. Voice Physiology and Biomechanics report. Marseille; page 137-143.

Hopkins 2006

Hopkins C, Pedersen M, Yousaf U. Acid Reflux Treatment for Hoarseness (Review). January 2006, Cochrane Library Issue 1.Wiley publishing.

Pedersen 2006

Pedersen M, Yousaf U. Videostroboscopic expert evaluation of the larynx with running objective voice measurement at the same time gives more secure results than videos alone, a case-control study.  Fifth International Conference on Voice Physiology and Biomechanics, 2006: 110-113.

Pedersen 2007

Pedersen M, Munck K. A prospective case-control study of jitter%, shimmer% and Qx%, glottis closure cohesion factor and Long Time Average Spectra. (Spead by Laryngograph Ltd.) Congress report, Models and analysis of vocal emissions for biomedical applications(MAVEBA), 2007; pages 60-64.

Aspects of consideration in database collection
Juan Ignacio Godino Llorente
In the context of the evaluation and assessment of speech disorders it is
very important to retrieve information of the patient along with the
multimedia data that supports the diagnosis (i.e. videoendoscopic,
electroglotographic, microphonic signal, etc). In this sense, this talk will
propose to the students a generic and structured framework for storing in a
PC the most important data for this purpose.
On the other hand, with respect to the recording of speech and
electroglottographic signals, this talk will pay attention to some problems
that usually arise making the recorded material invalid for further
processing: the selection and configuration of the ADC (analogue to digital
converter), the choice of the microphone, etc. 

Numerical modelling of voice function

Malte Kob

The function of human voice generation can be modelled using physical devices or computational methods. Numerical models have been developed for a number of applications and are used for simulation of various voice types including pathological voice generation. The concepts and implementations of such models are demonstrated. Perspectives and limitations of numerical voice modelling are discussed.

Mathematical Modeling of Speech Signals
Yannis Stylianou
Sinusoidal and harmonic plus noise speech models have been successfully applied in applications where coarse-resolution speech analysis is sufficient (i.e., speech coding) or where the stationarity assumption of speech is acceptable (i.e., speech synthesis). For voice quality assessment and voice
pathology detection based on the speech signal, however, a fine-resolution speech analysis is required while the dynamic characteristics of speech should be taken into account.
In this course we will present shortly the Sinusoidal and Harmonic models for speech as a necessary background information and then we will describe a  family of time-varying quasi-harmonic speech models, we will discuss analytically their properties and we will provide ways for the estimation of their parameters.

SHORT BIOS OF THE TEACHERS

Philippe Dejonckere

Philippe H. DeJonckere, M.D., Ph.D., is the chairman of COST-Action 2103, and chairman of the European Laryngological Research Group (ELRG) incorporated within the European Laryngological Society (ELS). This group of clinical researchers sought out a cooperation with engineers and physicists specialized in signal processing and modelling, and this resulted in COST Action 2103 "Advanced Voice Function Assessment". P. H. DeJonckere is ENT-physician / Phoniatrician, also certified specialist in Occupational Medicine, Rehabilitation Medicine and Forensic Medicine. He is currently full professor at the Utrecht University (NL), in charge of the Master programme "Clinical Language, Speech & Hearing Sciences", guest professor at the Catholic University of Leuven (B) and visiting professor at the University of Lille (F). He also is appointed as General Coordinator of the Scientific Council at the Federal Institute of Occupational Diseases in Brussels. He is past-president of the Collegium Medicorum Theatri and of the Dutch Society of Voice, Speech & Language Pathology, member of the editorial board of 11 scientific journals, and authored 7 books ,139 articles in international peer-reviewed journals and 69 chapters in books.

Ingo R. Titze

http://www.ncvs.org/ncvs/about/people/titze.htm

Jan Švec

Jan G. Svec, Ph.D. is an internationally renowned Czech physicist performing basic
research on production of human voice. He has worked as a research scientist at the
Center for Communication Disorders (Medical Healthcom) in Prague, the Czech Republic, at
the National Center for Voice and Speech in Denver, CO, USA and at the University of
Groningen, the Netherlands. Currently he is at the Palacky University Olomouc, the Czech
Republic. He designed videokymography, the method for high-speed visualization of
vocal-fold vibrations. He collaborates with numerous research teams in Europe and USA and
lectures world-wide. He is the current chairman of the Voice Committee of the
International Association of Logopedics and Phoniatrics (IALP).

Peter Murphy

Dr. Murphy received a BSc (Hons) in physics and mathematics from Trinity College, Dublin. He received a Ph.D. in physical sciences (speech analysis) from Dublin City University in 1997 following which he worked as a postdoctoral fellow, on voice analysis in the School of Control Systems and Electrical Engineering, Dublin Institute of Technology and School of Clinical Speech and Language Studies, Trinity College, Dublin. Dr. Murphy is currently a lecturer in Electronic Engineering, University of Limerick (since 1999).

He has sixteen years research expertise in the fields of voice science and voice processing for which he runs an active research program. He has a particular research interest in applying improvements in voice production knowledge to speech technology development. Over the past nine years he has attracted over €770,000 for ten funded projects (seven as principal investigator). He has published over 50 peer reviewed articles in scientific journals, book chapters and conference proceedings. Dr. Murphy is a member of the Acoustical Society of America, IEEE and the International Speech Communication Association. He is a member of the EU COST 2103 (Advanced Voice Function Assessment) and COST 2102 (Cross Modal Analysis of Verbal and Non-verbal Communication) Actions (2006-2010). He was a member of the EU COST 277 Action on Nonlinear Speech Processing (2001-2005).

Malte Kob

Dr.-Ing. Malte Kob
Department of Phoniatrics, Pedaudiology, and Communication Disorders
RWTH Aachen University, Aachen, GERMANY

Malte Kob studied electrical engineering at the Technical University of Braunschweig,
Germany with focus on signal processing, telecommunication engineering and semiconductor
physics. He received his doctoral degree in engineering sciences from RWTH Aachen
University, Germany, in 2002. Since 2001, he is head of the section `Voice and Hearing
Acoustics' of the Department of Phoniatrics, Pedaudiology, and Communication Disorders at
the University Hospital Aachen, Faculty of Medicine of RWTH Aachen University. He works
in teaching and research in the fields of medical acoustics, voice assessment &
classification and occupational voice disorders. He received the promotion award of the
German Society of Phoniatrics & Pedaudiology (DGPP) in 2003 and the Cremer award of the
German Acoustical Society (DEGA) in 2005. Malte Kob authored and coauthored more than 60
books, articles, research reports and conference presentations. He is member of the
Executive Council of the European Acoustics Association (EAA) and of the management
committee of COST action 2103 "Advanced Voice Function Assessment". He is a member of
ASA, DEGA and DPG and an ambitious Jazz musician (p,b).

Web pages:
http://www.phoniatrie.ukaachen.de/
http://www.akustik.rwth-aachen.de/~malte

Paavo Alku

Paavo Alku received the M.Sc, Lic.Tech, and Dr.Sc.(Tech) degrees from
Helsinki University of Technology, Espoo, Finland in 1986, 1988, and 1992,
respectively. He was Assistant Professor with the Asian Institute of Technology, Bangkok,
Thailand, in 1993, and the University of Turku, Finland, in 1994-1999. Currently,
he is professor of speech communication technology at Helsinki University
of Technology. His research interests are analysis and parameterization of speech
production, spectral modelling of speech, speech enhancement and cerebral processing of
speech.

See: http://www.acoustics.hut.fi/~paavo/

Jean Schoentgen has studied Engineering in Luxembourg and Physics at the Université Libre de Bruxelles, Brussels, Belgium, from which he holds a doctoral degree and a degree of "Agrégé de l'Enseignement Supérieur" (habilitation). Currently, he is a Senior Research Associate with the Fund for Scientific Research (FNRS) and a Lecturer in General and Clinical Phonetics at the Université Libre de Bruxelles and Université Catholique de Louvain. He is the secretary of COST Action 2103 on Advanced Voice Function Assessment and the chairman of the FNRS Contact Group on Speech and Language. His research interests include speech signal processing, voice quality, functional assessment of voice and speech as well as acoustic-to-articulatory mapping.

Mette Pedersen

http://www.mpedersen.org/

Juan Ignacio Godino Llorente was awarded an M.Sc. in Communications
Engineering in 1996 and a PhD. Degree with Honors in Computer Science from
the Universidad Politécnica de Madrid (UPM), Spain, in 2002. Currently he is
an Associate Professor at the UPM. Also, he is the Head of the Circuits and
Systems Engineering Department, which belongs to the same University. His
main research areas are in the field of biomedical signal processing, with
applications to voice disorders and ECG signal processing.
 
http://www.ics.upm.es/profesor.asp?codigo=igodino

Yannis Stylianou is Associate Professor at University of Crete, Department of Computer Science, CSD UOC and Associate Researcher in the Networks and Telecommunications Laboratory of the Institute of Computer Science ICS at FORTH. He received the Diploma of Electrical Engineering from the National Technical University, N.T.U.A., of Athens in 1991 and the M.Sc. and Ph.D. degrees in Signal Processing from the Ecole National Superieure des Telecommunications, ENST, Paris, France in 1992 and 1996, respectively. From 1996 until 2001 he was with AT&T Labs Research (Murray Hill and Florham Park, NJ, USA) as a Senior Technical Staff Member. In 2001 he joined Bell-Labs Lucent Technologies, in Murray Hill, NJ, USA (now Alcatel-Lucent). Since 2002 he is with the Computer Science Department at the University of Crete and the Institute of Computer Science at FORTH.

He is member of the IEEE Speech and Language Technical Committee and Associate Editor of the EURASIP Journal on Speech, Audio, and Music Processing, ASMP, and of the EURASIP Research Letters in Signal Processing, RLSP. He is Vice-Chairman of the Cost Action 2103: "Advanced Voice Function Assessment", VOICE. He was Associate Editor for the IEEE Signal Processing Letters and on the Management Committee for the COST Action 277: "Nonlinear Speech Processing". He holds 9 patents and he is member of IEEE and IEEE SPS, and of the Technical Chamber of Greece, TEE.

His current research focuses on speech signal processing algorithms for speech analysis, statistical signal processing (detection and estimation), and time-series analysis/modeling.

During Interspeech 2007, Antwerp, Belgium he gave a Tutorial on Voice Transformation.