Invited Papers -- Speech and Language Processing: Can We Use the Past to Predict the Future? -- Common Sense About Word Meaning: Sense in Context -- ScanSoft’s Technologies -- Text -- A Positional Linguistics-Based System for Word Alignment -- Handling Multi-word Expressions Without Explicit Linguistic Rules in an MT System -- The Szeged Corpus: A POS Tagged and Syntactically Annotated Hungarian Natural Language Corpus -- Item Summarization in Personalisation of News Delivery Systems -- IR-n System, a Passage Retrieval Architecture -- Event Clustering in the News Domain -- HANDY: Sign Language Synthesis from Sublexical Elements Based on an XML Data Representation -- Using Linguistic Resources to Construct Conceptual Graph Representation of Texts -- Slovak National Corpus -- Grammatical Heads Optimized for Parsing and Their Comparison with Linguistic Intuition -- How Dominant Is the Commonest Sense of a Word? -- POS Tagging of Hungarian with Combined Statistical and Rule-Based Methods -- Grammatical Relations Identification of Korean Parsed Texts Using Support Vector Machines -- Clustering Abstracts Instead of Full Texts -- Bayesian Reinforcement for a Probabilistic Neural Net Part-of-Speech Tagger -- Automatic Language Identification Using Phoneme and Automatically Derived Unit Strings -- Slovak Text-to-Speech Synthesis in ARTIC System -- Identifying Semantic Roles Using Maximum Entropy Models -- A Lexical Grammatical Implementation of Affect -- Towards Full Lexical Recognition -- Discriminative Models of SCFG and STSG -- Coupling Grammar and Knowledge Base: Range Concatenation Grammars and Description Logics -- Balancing Manual and Automatic Indexing for Retrieval of Paper Abstracts -- Unsupervised Learning of Rules for Morphological Disambiguation -- Ambiguous Supertagging Using a Feature Structure -- A Practical Word Sense Disambiguation System with High Performance for Korean -- Morphological Tagging of Russian Texts of the XIXth Century -- Speech -- Large Vocabulary Continuous Speech Recognition for Estonian Using Morphemes and Classes -- A New Classifier for Speaker Verification Based on the Fractional Brownian Motion Process -- A Universal Human Machine Speech Interaction Language for Robust Speech Recognition Applications -- Embedded ViaVoice -- New Speech Enhancement Approach for Formant Evolution Detection -- Measurement of Complementarity of Recognition Systems -- Text-to-Speech for Slovak Language -- Speaker Verification Based on Wavelet Packets -- A Decoding Algorithm for Speech Input Statistical Translation -- Aggregation Operators and Hypothesis Space Reductions in Speech Recognition -- Combinations of TRAP Based Systems -- Automatic Recognition and Evaluation of Tracheoesophageal Speech -- Using Neural Networks to Model Prosody in Czech TTS System Epos -- Auditory Scene Analysis via Application of ICA in a Time-Frequency Domain -- Using the Lemmatization Technique for Phonetic Transcription in Text-to-Speech System -- Automatic Categorization of Voicemail Transcripts Using Stochastic Language Models -- Low Latency Real-Time Vocal Tract Length Normalization -- Multimodal Phoneme Recognition of Meeting Data -- A New Multi-modal Database for Developing Speech Recognition Systems for an Assistive Technology Application -- Obtaining and Evaluating an Emotional Database for Prosody Modelling in Standard Basque -- Fully Automated Approach to Broadcast News Transcription in Czech Language -- A Computational Model of Intonation for Yorùbá Text-to-Speech Synthesis: Design and Analysis -- Dynamic Unit Selection for Very Low Bit Rate Coding at 500 bits/sec -- On the Background Model Construction for Speaker Verification Using GMM -- A Speaker Clustering Algorithm for Fast Speaker Adaptation in Continuous Speech Recognition -- Advanced Prosody Modelling -- Voice Stress Analysis -- Slovak Speech Database for Experiments and Application Building in Unit-Selection Speech Synthesis -- Towards Lower Error Rates in Phoneme Recognition -- Examination of Pronunciation Variation from Hand-Labelled Corpora -- New Refinement Schemes for Voice Conversion -- Acoustic and Linguistic Information Based Chinese Prosodic Boundary Labelling -- F0 Prediction Model of Speech Synthesis Based on Template and Statistical Method -- An Architecture for Spoken Document Retrieval -- Evaluation of the Slovenian HMM-Based Speech Synthesis System -- Modeling Prosodic Structures in Linguistically Enriched Environments -- Parallel Root-Finding Method for LPC Analysis of Speech -- Automatic General Letter-to-Sound Rules Generation for German Text-to-Speech System -- Pitch Accent Prediction from ToBI Annotated Corpora Based on Bayesian Learning -- Processing of Logical Expressions for Visually Impaired Users -- Dialogue -- Durational Aspects of Turn-Taking in Spontaneous Face-to-Face and Telephone Dialogues -- A Speech Platform for a Bilingual City Information System -- Rapid Dialogue Prototyping Methodology -- Building Voice Applications from Web Content -- Information-Providing Dialogue Management -- Realistic Face Animation for a Czech Talking Head -- Evaluation of a Web Based Information System for Blind and Visually Impaired Students: A Descriptive Study -- Multimodal Dialogue Management -- Looking at the Last Two Turns, I’d Say This Dialogue Is Doomed – Measuring Dialogue Success -- Logical Approach to Natural Language Understanding in a Spoken Dialogue System -- Building a Dependency-Based Grammar for Parsing Informal Mathematical Discourse.

This volume contains the Proceedings of the 7th International Conference on Text, Speech and Dialogue, held in Brno, Czech Republic, in September 2004, under the auspices of the Masaryk University. This series of international conferences on text, speech and dialogue has come to c- stitute a major forum for presentation and discussion, not only of the latest developments in academic research in these ?elds, but also of practical and industrial applications. Uniquely, these conferences bring together researchers from a very wide area, both intellectually and geographically, including scientists working in speech technology, dialogue systems, text processing, lexicography, and other related ?elds. In recent years the conference has dev- oped into aprimary meetingplacefor speech and languagetechnologistsfrom manydifferent parts of the world and in particular it has enabled important and fruitful exchanges of ideas between Western and Eastern Europe. TSD 2004 offered a rich program of invited talks, tutorials, technical papers and poster sessions, aswellasworkshops andsystemdemonstrations. Atotalof78paperswereaccepted out of 127 submitted, contributed altogether by 190 authors from 26 countries. Our thanks as usual go to the Program Committee members and to the external reviewers for their conscientious and diligent assessment of submissions, and to the authors themselves for their high-quality contributions. We would also like to take this opportunity to express our appreciation to all the members of the Organizing Committee for their tireless efforts in organizing the conference and ensuring its smooth running.