27th International Conference on Text, Speech and Dialogue
TSD 2024, Brno, Czech Republic, September 9–13 2024
 
Topics | Committees | Important Dates | Contact info | Call for Papers
Conference Program

For technical details see the equipment list.

Overview
DateTimeActionRoom
Tuesday, September 10, 202416:00 - 18:00RegistrationLounge
Wednesday, September 11, 20248:00 - 9:00RegistrationLounge
9:00 - 9:20Opening SessionHall 5
9:20 - 10:20Hynek Hermansky:
Why should we ask why? - invited talk
Hall 5
10:20 - 10:35Coffee BreakLounge
10:35 - 12:15Parallel Sessions (2 x 4)Hall 5, Hall 4
12:15 - 13:30Lunch BreakRestaurant
13:30 - 14:30Poster SessionLounge
14:30 - 16:10Parallel Sessions (2 x 4)Hall 5, Hall 4
16:10 - 16:25Coffee BreakLounge
16:25 - 17:30Demo SessionHall 4
16:30 - 17:30Program Committee MeetingHall 5
18:00 - 20:00Welcome ReceptionRestaurant
Thursday, September 12, 202409:00 - 10:15Parallel Sessions (2 x 3)Hall 5, Hall 4
10:15 - 10:30Coffee BreakLounge
10:30 - 11:45Parallel Sessions (2 x 3)Hall 5, Hall 4
11:45 - 12:45Lunch BreakRestaurant
12:45 - 23:00Trip and Conference Dinner
Friday, September 13, 20249:30 - 10:30Preslav Nakov:
Factuality Challenges in the Era of Large Language Models: Can we Keep LLMs Safe and Factual? - invited talk
Hall 5
10:30 - 10:45Coffee BreakLounge
10:45 - 12:25Parallel Sessions (2 x 4)Hall 5, Hall 4
12:25 - 12:30Closing CeremonyHall 5
12:30 - 13:30Lunch BreakRestaurant










Wednesday, September 11, 2024
Time
8:00Registration (Lounge)
9:00Opening Session (Hall 5)
9:20Hynek Hermansky:
Why should we ask why? - invited talk
(Hall 5)
chair: Elmar Nöth
10:20Coffee Break (Lounge)
Parallel Sessions (2 x 4)
Section Text (Hall 5)
chair: Pavel Rychlý
Section Speech (Hall 4)
chair: Jindřich Matoušek
10:35#1202: Evangelia Gogoulou and Timothée Lesort and Magnus Boman and Joakim Nivre:
Continual Learning Under Language Shift (oral)
#1291: Maroš Jakubec and Roman Jarina and Eva Lieskovská and Peter Kasák and Michal Spišiak:
Enhancing Speech Emotion Recognition Using Transfer Learning From Speaker Embeddings (oral)
11:00#1196: Jérémie Roux and Hani Guenoune and Mathieu Lafourcade and Richard Moot:
Explaining Metaphors in the French Language by Solving Analogies using a Knowledge Graph (oral)
#1215: Mohammed Hamzah Abed and Dávid Sztahó:
Deep Speaker Embeddings for Speaker Verification of Children (oral)
11:25#1252: Marisa Schmidt and Karin Harbusch and Denis Memmesheimer:
Automatic Ellipsis Reconstruction in Coordinated German Sentences Based on Text-To-Text Transfer Transformers (oral)
#1218: Chin Yuen Kwok and Jia Qi Yip and Eng Siong Chng:
Improved Alignment for Score Combination of RNN-T and CTC Decoder for Online Decoding (oral)
11:50#1269: Michael Mohler and Sandra Lee and Mary Brunson and David Bracewell:
Introducing LCC's NavProc 1.0 Corpus (oral)
#1285: Mala J. B. and Alex Raj S. M. and Rajeev Rajan:
X-vector-based Speaker Diarization Using Bi-LSTM and Interim Voting-driven Post-processing (oral)
12:15Lunch Break (Restaurant)
13:30Poster Session (Lounge)
chair: Ales Horak
#1211: Krištof Anetta and Aleš Horák:
New Human-Annotated Dataset of Czech Health Records for Training Medical Concept Recognition Models (poster)
#1225: Chang Nian Chuy and Cherie Ding and Qinmin Vivian Hu:
Analyzing Biases in Popular Answer Selection Datasets on Neural-based QA Models (poster)
#1231: Lilia Azrou and Houda Oufaida and Philippe Blache and Israa Hamdine:
Using Neural Coherence Models to Assess Discourse Coherence (poster)
#1253: Edoardo Signoroni and Pavel Rychlý:
Better Low-Resource Machine Translation with Smaller Vocabularies (poster)
#1264: Adam Mištera and Tomáš Brychcín:
Kernel Least Squares Transformations for Cross-lingual Semantic Spaces (poster)
#1267: Abishek Stephen and Vojtěch John and Zdeněk Žabokrtský:
Unsupervised Extraction of Morphological Categories for Morphemes (poster)
#1200: Gokul Srinivasagan and Munir Georges:
Retrieval Augmented Spoken Language Generation for Transport Domain (poster)
#1205: Sven Aller and Mark Fishel:
Adapting Audiovisual Speech Synthesis to Estonian (poster)
#1206: Dosti Aziz and Dávid Sztahó:
Dysphonia Diagnosis Using Self-Supervised Speech Models in Mono- and Cross-Lingual Settings (poster)
#1261: David Porteš and Aleš Horák:
Generating High-Quality F0 Embeddings Using the Vector-Quantized Variational Autoencoder (poster)
#1273: Abner Hernandez and Paula Andrea Perez-Toro and Tomas Arias-Vergara and Juan Camilo Vasquez-Correa and Seung Hee Yang and Juan Rafael Orozco-Arroyave and Andreas Maier:
Anonymizing Dysarthric Speech: Investigating the Effects of Voice Conversion on Pathological Information Preservation (poster)
#1233: Ondřej Sotolář and Jaromír Plhák and David Šmahel:
Leveraging Conceptual Similarities to Enhance Modeling of Factors Affecting Adolescents' Well-Being (poster)
#1242: Niko Kleer and Leon Weyand and Michael Feld and Klaus Berberich:
Capturing Task-Related Information for Text-Based Grasp Classification Using Fine-Tuned Embeddings (poster)
#1268: Julian Wolter and Niko Kleer and Michael Feld:
StepDP: A Step Towards Expressive and Pervasive Dialogue Platforms (poster)
Parallel Sessions (2 x 4)
Section Text (Hall 5)
chair: Karin Harbusch
Section Dialogue (Hall 4)
chair: Daniel Tihelka
14:30#1210: Julien Delaunay and Hanh Thi Hong Tran and Carlos-Emiliano González-Gallardo and Georgeta Bordea and Mathilde Ducos and Nicolas Sidere and Antoine Doucet and Senja Pollak and Olivier De Viron:
CoastTerm: a Corpus for Multidisciplinary Term Extraction in Coastal Scientific Literature (oral)
#1222: Diego Alexander Lopez-Santander and Cristian David Rios-Urrego and Christian Bergler and Elmar Nöth and Juan Rafael Orozco-Arroyave:
Robust Classification of Parkinson’s Speech: an Approximation to a Scenario With Non-controlled Acoustic Conditions (oral)
14:55#1197: Vladimír Benko:
The Aranea Corpora Family: Ten+ Years of Processing Web-Crawled Data (oral)
#1238: Ankit Kumar and Munir Georges:
Joint-Average Mean and Variance Feature Matching (JAMVFM) Semi-supervised GAN with Additional-Objective Training Function for Intent Detection (oral)
15:20#1256: Duygu Altinok:
Bella Turca: A Large-Scale Dataset of Diverse Text Sources for Turkish Language Modeling (oral)
#1203: Lucas Druart and Valentin Vielzeuf and Yannick Estève:
Investigating Low-Cost LLM Annotation for Spoken Dialogue Understanding Datasets (oral)
15:45#1282: Anastasiia Aleksandrova and Joakim Nivre:
Models and Strategies for Russian Word Sense Disambiguation: A Comparative Analysis (oral)
#1216: Daniel Ortega and Steven Söhnel and Ngoc Thang Vu:
Improving and Understanding Clarifying Question Generation in Conversational Search (oral)
16:10Coffee Break (Lounge)
16:25Demo Session (Hall 4)
chair: Zuzana Nevěřilová
#1270: Di Wu, Munir Georges:
A Pipeline for Automatic Construction and Applications of College Curriculum Knowledge Graph (demo)
#1300: Hana Žižková:
Opravidlo: From Beta version to Opravidlo2.0 (demo)
#1301: Martin Fajcik:
A Leaderboard for BenCzechMark: A Czech-centric Multitask and Multimetric Benchmark for Language Models with Duel Scoring Mechanism (demo)
#1280: DigiEduBerlin Team:
Greet the speaking book: Pupil Identification on Personal Primer Prototypes 1, 2, 4, 8 (demo)
#1302: Michal Cukr:
OneClick Terms: Bilingual Terminology Extraction (demo)
#1303: Zuzana Nevěřilová:
Language Services (demo)
#1304: Ondřej Herman:
Unsupervised Sense Classification For Word Sketches (demo)
17:30
16:30Program Committee Meeting (Hall 5)
17:30
18:00Welcome Reception (Restaurant)
20:00










Thursday, September 12, 2024
TimeParallel Sessions (2 x 3)
Section Text/Speech (Hall 5)
chair: Vladimír Benko
Section Speech (Hall 4)
chair: Juan Rafael Orozco-Arroyave
09:00#1192: Hanh Thi Hong Tran and Carlos-Emiliano González-Gallardo and Julien Delaunay and Antoine Doucet and Senja Pollak:
Is Prompting What Term Extraction Needs? (oral)
#1244: Zdeněk Hanzlíček:
Data Alignment and Duration Modelling in VITS (oral)
09:25#1246: Zuzana Nevěřilová and Hana Žižková:
Named Entity Linking in English-Czech Parallel Corpus (oral)
#1219: Erfan A. Shams and Julie Carson-Berndsen:
Attention to Phonetics: A Visually Informed Explanation of Speech Transformers (oral)
09:50#1247: Ilaria Manfredi:
Multiword Expressions Resources for Italian: Presenting a Manually Annotated Spoken Corpus (oral)
#1208: Daniel Tihelka and Jindřich Matoušek and Zdeněk Hanzlíček and Lukáš Vladař:
Sentences vs Phrases in Neural Speech Synthesis (oral)
10:15Coffee Break (Lounge)
Parallel Sessions (2 x 3)
Section Text (Hall 5)
chair: Zuzana Nevěřilová
Section Dialogue (Hall 4)
chair: Jan Lehečka
10:30#1262: Kai Hartung and Sambit Mallick and Sören Gröttrup and Munir Georges:
Evaluation Metrics in LLM Code Generation (oral)
#1212: Kwan-yeung Wong and Fu-lai Chung:
PiCo-VITS: Leveraging Pitch Contours for Fine-grained Emotional Speech Synthesis (oral)
10:55#1293: Mária Pappová and Matúš Valko:
Mistrík's Readability Metric - an Online Library (oral)
#1271: Jeferson David Gallo-Aristizábal and Daniel Escobar-Grisales and Cristian David Ríos-Urrego and Elmar Nöth and Juan Rafael Orozco-Arroyave:
Automatic Classification of Parkinson's Disease Using Wav2vec Embeddings at Phoneme, Syllable, and Word Levels (oral)
11:20#1251: Randil Pushpananda and Chamila Liyanage and Ashmari Pramodya and Ruvan Weerasinghe:
TamSiPara: A Tamil - Sinhala Parallel Corpus (oral)
#1217: Duygu Altinok:
Explainable Multimodal Fusion for Dementia Detection from Text and Speech (oral)
11:45Lunch Break (Restaurant)
12:45Trip and Conference Dinner
23:00










Friday, September 13, 2024
Time
9:30Preslav Nakov:
Factuality Challenges in the Era of Large Language Models: Can we Keep LLMs Safe and Factual? - invited talk
(Hall 5)
chair: Ales Horak
10:30Coffee Break (Lounge)
Parallel Sessions (2 x 4)
Section Text (Hall 5)
chair: Petr Sojka
Section Speech (Hall 4)
chair: Zdeněk Hanzlíček
10:45#1287: Milan Straka and Jana Straková:
Open-Source Web Service with Morphological Dictionary-Supplemented Deep Learning for Morphosyntactic Analysis of Czech (oral)
#1213: Jan Lehečka and Zdeněk Hanzlíček and Jindřich Matoušek and Daniel Tihelka:
Zero-Shot vs. Few-Shot Multi-Speaker TTS Using Pre-trained Czech SpeechT5 Model (oral)
11:10#1209: Matej Klemen and Martin Božič and Špela Arhar Holdt and Marko Robnik-Šikonja:
Neural Spell-Checker: Beyond Words with Synthetic Data Generation (oral)
#1241: Santiago A. Moreno-Acevedo and Juan Camilo Vasquez-Correa and Juan M. Martín-Doñas and Aitor Álvarez:
Stream-Based Active Learning for Speech Emotion Recognition via Hybrid Data Selection and Continuous Learning (oral)
11:35#1195: Michaela Denisová and Pavel Rychlý:
Bilingual Lexicon Induction From Comparable and Parallel Data: A Comparative Analysis (oral)
#1289: Thibault Bañeras-Roux and Mickael Rouvier and Jane Wottawa and Richard Dufour:
A Paradigm for Interpreting Metrics and Measuring Error Severity in Automatic Speech Recognition (oral)
12:00#1191: Maixent Chenebaux and Tristan Cazenave:
SeqCondenser: Inductive Representation Learning of Sequences by Sampling Characteristic Functions (oral)
#1234: Lukáš Vladař and Jindřich Matoušek:
Effects of Training Strategies and the Amount of Speech Data on the Quality of Speech Synthesis (oral)
12:25Closing Ceremony (Hall 5)
12:30Lunch Break (Restaurant)
13:30










.
TSD 2023 | TSD 2022 | TSD 2021