3rd Graz-Vienna Speech Workshop

Room: Hörsaal 5

• 12:00 - 13:00 hrs

  Registration

• 13:00 - 13:10 hrs

  Welcome by the organizers

Room: Hörsaal 5

Chair: Veronika Mattes (Department of Linguistics, University of Graz)

• 13:10 - 13:30 hrs

  On the Vowel Development of Bilingual Kindergarten Children in their L2 German

  Carolin Schmid (Department of Pediatrics and Adolescent Medicine, Medical University of Vienna)

• 13:30 - 13:50 hrs

  Expressive Vocabulary Development in Children with Permanent Hearing Loss: Early Nonverbal Predictors

  Anna-Lena Feichter (Department of Linguistics, University of Graz)

• 13:50 - 14:10 hrs

  Phonetic Analysis of Dysarthric Child Speech

  Martina Galović (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 14:10 - 14:30 hrs

  Towards Developing an Assistive Communication Tool for Dysarthric Children

  Barbara Schuppler (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 14:30 - 14:50 hrs

  Multi-Style Child TTS: An Expressive Text-to-Speech System for Children

  Shaima Alwaisi (Department of Telecommunications and Artificial Intelligence, Budapest University of Technology and Economics)

• 14:50 - 15:10 hrs

  Wrap up & discussion

Room: Foyer, Level 8

• 14:50 - 15:10 hrs

  Coffee break

Room: Hörsaal 5

Chair: Barbara Schuppler (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 15:40 - 16:00 hrs

  An Acoustic and Articulatory Study of Voice Quality in Coarticulated Hanoi Vietnamese Tones

  Yaqian Huang (Acoustics Research Institute, Austrian Academy of Science)

• 16:00 - 16:20 hrs

  Does Spatial Auditory Attention Change After Having a Face-To-Face Conversation in a Noisy Environment?

  Ľuboš Hládek (Acoustics Research Institute, Austrian Academy of Science)

• 16:20 - 16:40 hrs

  (When) Does it Harm to Be Incomplete? Encoding ASR Mistranscriptions of Syntactically Disfluent Structures

  Saskia Wepner (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 16:40 - 17:00 hrs

  ASR of Non-lexical and Disfluent Events -an Investigation Across Tasks

  Yan Meng (Department of Telecommunications and Artificial Intelligence, Budapest University of Technology and Economics)

• 17:00 - 17:20 hrs

  Speech Enhancement of Conversational Speech in Cocktail Party Noise

  Emil Berger (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 17:20 - 17:40 hrs

  Building a Robust HMI Command Inventory for Various Noisy Environments

  Sara Marguerite Pearsell (Institute of Mechanical and Electrical Engineering, University of Southern Denmark)

• 17:40 - 18:00 hrs

  Wrap up & Discussion

Room: Hörsaal 5

Chair: Peter Mihajlik (Department of Telecommunications and Artificial Intelligence, Budapest University of Technology and Economics)

• 09:00 - 09:20 hrs

  Comparison of Self-Training Strategies for ASR

  Yue Luo (Department of Telecommunications and Artificial Intelligence, Budapest University of Technology and Economics)

• 09:20 - 09:40 hrs

  Improving Khalkha Mongolian ASR via Transliteration-Based Transfer Learning

  Dalai Mengke (Department of Telecommunications and Artificial Intelligence, Budapest University of Technology and Economics)

• 09:40 - 10:00 hrs

  Speech Event Extraction: An ASR+NLP Challenge

  Máté Gedeon (Department of Telecommunications and Artificial Intelligence, Budapest University of Technology and Economics)

• 10:00 - 10:20 hrs

  Towards Creating a Carinthian Slovenian Spoken Language Resource

  Andrej Žgank (Faculty of Electrical Engineering and Computer Science, University of Maribor)

• 10:20 - 10:40 hrs

  Wrap up & Discussion

• 10:40 hrs

  Group photo

Room: Foyer, Level 8

• 10:50 - 11:10 hrs

  Coffee break

Room: Kursraum 21

• 11:10 - 11:55 hrs

  How to Use SLICER for Stimuli Extraction from Large Speech Corpora

  Lucas Eckert (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 12:00 - 12:45 hrs

  Annotating Creak in Healthy and Pathological Voices

  Anna Viehauser (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

Room: Mensa/Personalspeiseraum, Level 5

• 12:45 - 14:00 hrs

  Lunch break

Room: Kursraum 21

• 14:00 - 14:45 hrs

  SpeechScape: An Easy-to-Use Tool for Clustering Speech Data Based on Self-Supervised Representations

  Julian Linke (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 14:45 - 15:35 hrs

  Speech Pauses in Alzheimer Disease: Exploring Challenges in Training Automatic Speech Recognition Systems

  Sandra Lafenthaler (Department of Neurology, Neurointensive Care, and Neurorehabilitation, Paracelsus Medical University Salzburg)

• 15:40 - 16:25 hrs

  From Medical Data to Models: Own Experiences and Challenges

  Mario Fleischer (Department of Audiology and Phoniatrics, Charité – Universitätsmedizin Berlin)

Room: all rooms (Hörsaal 5, Kursräume 21-23)

• 16:30 - 17:30 hrs

  Discussion and exchange in small self-organized groups

• 11:10 - 12:45 hrs

  Evidencing Physiological and Acoustic Outcomes of Clinical Voice Interventions Using Voice Maps

  Sten Ternström (Department of Speech, Music and Hearing, KTH Royal Institute of Technology, Stockholm)

Room: Mensa/Personalspeiseraum, Level 5

• 12:45 - 14:00 hrs

  Lunch break

Room: Kursraum 22

• 14:00 - 14:45 hrs

  Optopalatography for Phonetic Research

  Jihyeon Yun (Institute of Acoustics and Speech Communication, Dresden University of Technology)

• 14:50 - 15:35 hrs

  Using Ultrasound Tongue Imaging for Phonetic Research

  Nathalie Elsässer (Acoustics Research Institute, Austrian Academy of Sciences)

• 15:40 - 16:25 hrs

  Analysis of Uni- and Multi-Dimensional Contours with GAMs and Functional PCA: an Application to Ultrasound Tongue Imaging

  Michele Gubian (Institute for Phonetics and Speech Processing, LMU Munich)

Room: all rooms (Hörsaal 5, Kursräume 21-23)

• 16:30 - 17:30 hrs

  Discussion and exchange in small self-organized groups

Room: Hörsaal 5

Chair: Philipp Aichinger (Department of Otorhinolaryngology, Medical University of Vienna)

• 09:00 - 09:20 hrs

  Project Kick-Off: Voice Conversion for Pathological Speech

  Martin Hagmüller (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

  Philipp Aichinger (Department of Otorhinolaryngology, Medical University of Vienna)

• 09:20 - 09:40 hrs

  Voice Conversion in Pathological Speech: Applications and Challenges

  Benedikt Mayrhofer (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 09:40 - 10:00 hrs

  Realtime Personalized Deep Learning Voice Morphing for Electrolaryngeal Speech in Polish Language

  Pawel Cyrta (Uhura Bionics Ltd., Warsaw)

• 10:00 - 10:20 hrs

  A Medical Application of ASR and LLM

  Imre Gergely Jánoki (Department of Telecommunications and Artificial Intelligence,  Budapest University of Technology and Economics)

• 10:20 - 10:40 hrs

  The Mere-Measurement Effect in Patient-Reported Outcomes: A Randomized Controlled Trial with Speech Pathology Patients

  Preston Long (Centre for Medical Data Science, Medical University of Vienna)

Room: Foyer, Level 8

• 10:40 - 11:00 hrs

  Coffee break

Room: Kursraum 21

• 11:00 - 11:45 hrs

  Speech-Based Cardiorespiratory Health Monitoring with VOICE-BIOME: a scalable voice biomarker platform

  Prashanth Pombala (ZANA Technologies GmbH, Karlsruhe)

Room: Kursraum 22

• 11:00 - 11:45 hrs

  Deep Learning ASR for a Patient with Permanent Tracheostomy

  David Nadrchal (Institute of Computational Perception, Johannes Kepler University Linz)

Room: Hörsaal 5

Chair: Gernot Kubin (Signal Processing and Speech Communication Laboratory, Graz University of Technology)

• 11:50 - 12:10 hrs

  Physically-Based Machine Learning for Vocal Fold Video Data Interpretation

  Carlo Crioli (Dept. of Mathematics, Computer Science, and Physics DMIF, University of Udine)

• 12:10 - 12:30 hrs

  Sampling Rate Bias of Vocal Jitter and Shimmer

  Jean Schoentgen (Laboratories of Image, Signal Processing and Acoustics, Université Libre de Bruxelles)

• 12:30 - 12:50 hrs

  Resonance Frequencies in Non-Rigid Compressed Waveguides

  Annemie Van Hirtum (Laboratoire des Écoulements Géophysiques et Industriels LEGI, University of Grenoble Alpes)

• 12:50 - 13:10 hrs

  Physical Model of Phonation with Reduced and Measurable Parameters

  Xavier Pelorson (Laboratoire des Écoulements Géophysiques et Industriels LEGI, University of Grenoble Alpes)

• 13:10 - 13:30 hrs

  Wrap up & Discussion

Room: Mensa/Personalspeiseraum, Level 5

• 13:30 - 14:30 hrs

  Farewell lunch