Abstract not found

Generation of Intensive Care data) [1], an experimental system that we have been developing to produce briefings automatically of patient status after CABG (coronary artery bypass graft).

Abstract. A significant challenge in Text-to-Speech (TtS) synthesis is the formulation of the prosodic structures (phrase breaks, pitch accents, phrase accents and boundary tones) of utterances. The prediction of these elements robustly relies on the accuracy and the quality of error-prone linguistic procedures, such as the identification of the part-of-speech and the syntactic tree. Additional linguistic factors, such as rhetorical relations, improve the naturalness of the prosody, but are hard to extract from plain texts. In this work, we are proposing a method to generate enhanced prosodic events for TtS by utilizing accurate, error-free and high-level linguistic information. We are also presenting an appropriate XML annotation scheme to encode syntax, grammar, new or given information, phrase subject/object information, as well as rhetorical elements. These linguistically enriched has have been utilized to build realistic machine learning models for the prediction of the prosodic structures in terms of segmental information and ToBI marks. The methodology has been applied by exploiting a Natural Language Generator (NLG) system. The trained models have been built using classification via regression trees and the results strongly indicate the realistic effect on the generated prosody. The evaluation of this approach has been made by comparing the models produced by the enriched documents to those produced by plain text of the same domain. The results show an improved accuracy of up to 23%. 1.

The main focus of this thesis is on improving the quality of concatenative speech synthesis by taking advantage of the natural (allowable) variability in spoken language, namely, the fact that there are multiple ways of uttering a given sentence and there are several word sequences that can represent a given concept. An architecture for speech generation for constrained domain applications is proposed that tightly integrates language generation and speech synthesis, allowing the choice of words and desired intonation in the system's response to be optimized jointly with the speech output quality. Experiments with a travel planning dialog system have demonstrated that by expanding the space of candidate responses and possible prosodic realizations we achieve higher quality speech output.

this paper has been partially supported by the GR-PROSODY project of the KAPODISTRIAS Program, Special Account for Research Grants, National and Kapodistrian University of Athens and by the HERACLITUS project of the Operational Programme for Education and Initial Vocational Training (EPEAEK), Greek Ministry of Education, under the 3rd European Community Support Framework for Greece

This dissertation identifies and resolves constraints related to the task of combining related clauses to formulate fluent and concise sentences. To incorporate complex linguistic constructions into text generation systems, novel algorithms were designed to systematically generate conjunction, ellipsis, and quantification constructions. Casper a submodule in a text generation system, was designed and implemented. It can convey the same information using fewer words by taking advantage of redundancies in the input based on syntactic, semantic, and discourse information. In addition to these symbol approaches, my research also employs corpus-based statistical approaches to enhance the fluency of the generated text. By employing advance linguistic constructions and removing redundancies through clause aggregations, the generated text or speech is more fluent and concise and thus improves human-computer interface.

this paper has been partially supported by the HERACLITUS project of the Operational Programme for Education and Initial Vocational Training (EPEAEK) of the Greek Ministry of Education under the 3rd European Community Support Framework for Greece

Prosody, or intonation, is a critically important component of spoken communication. The automatic extraction of prosodic information is necessary for machines to process speech with human levels of proficiency. In this thesis we describe work on the automatic detection and classification of prosodic events – specifically, pitch accents and prosodic phrase boundaries. We present novel techniques, feature representations and state of the art performance in each of these tasks. We also present three proof-of-concept applications – speech summarization, story segmentation and non-native speech assessment – showing that access to hypothesized prosodic event information can be used to improve the performance of downstream spoken language processing tasks. We believe the contributions of this thesis advance the understanding of prosodic events and the use of prosody in spoken language processing towards the goal of human-like processing of speech by machines.

In this paper, we report on an experiment showing how the introduction of prosodic information from detailed syntactic structures into synthetic speech leads to better disambiguation of structurally ambiguous sentences. Using modifier attachment (MA) ambiguities and subject/object fronting (OF) in German as test cases, we show that prosody which is automatically generated from deep syntactic information provided by an HPSG generator can lead to considerable disambiguation effects, and can even override a strong semantics-driven bias. The architecture used in the experiment, consisting of the LKB generator running a large-scale grammar for German, a syntaxprosody interface module, and the speech synthesis system MARY is shown to be a valuable platform for testing hypotheses in intonation studies.