This paper proposes a way of preparing and recording a speech corpus for unit selection text-to-speech speech synthesis driven by symbolic prosody. The research is focused on a phonetically and prosodically rich sentence selection algorithm. Symbolic description on a deep prosody level ...

The paper gives a brief summarisation of preparation and recording of a phonetically and prosodically rich speech corpus for Czech unit selection text-to-speech synthesis. Special attention is paid to the process of two-phase orthographic annotations of recorded sentences with regard to thei...

This paper deals with an HMM-based automatic phonetic segmentation (APS) system and proposes to increase its performance by employing a pitch-synchronous (PS) coding scheme. Such a coding scheme uses different frames of speech throughout voiced and unvoiced speech regions and enables thus&#x...

The paper deals with the process of designing a phonetically and prosodically rich speech corpus for unit selection speech synthesis. The attention is given mainly to the recording and verification stage of the process. In order to ensure as high quality and consistency of the...

This paper presents initial experiments with the identification and automatic detection of parasitic sounds in speech signals. The main goal of this study is to identify such sounds in the source recordings for unit-selection-based speech synthesis systems and thus to avoid their uninte...

ARTIC (Artificial Talker in Czech) is a corpus-based text-to-speech (TTS) system that enables to synthesise an arbitrary text, mainly for the Czech language. Basically, two versions of ARTIC are available - a single unit instance system (also known as fixed-inventory synthesis) with the...

ARTIC (Artificial Talker in Czech) is a corpus-based text-to-speech (TTS) system that enables to synthesise an arbitrary text, mainly for the Czech language. Basically, two versions of ARTIC are available - a single unit instance system (also known as fixed-inventory synthesis) with the...

The paper deals with the dependence between the speaker identification performance and the amount of test data. Three speaker identification procedures based on hidden Markov models (HMMs) of phonemes are presented here. One, which is quite commonly used in the speaker recognition syste...

In this paper, three speaker verification procedures are tested. All the procedures are based on Gaussian mixture models (GMM), however, they differ in the way, in which they use particular feature vectors of an utterance for speaker verification. A lot of experiments have been...

A method of speaker verification based on Gaussian mixture models is presented in this paper. The method works with a background model which is composed of several submodels. Several different approaches for construction of the background model from the submodels are introduced here:&#x...

Our paper introduces implementation details of the application that serves as an audiovisual interface to the automatic dialogue system. It comprises a state-of-the-art large vocabulary continuous speech recognition engine and a TTS system coupled with an embodied avatar that is able to ...

The paper presents two different methods for adding previously unseen words into the LVCSR system. Both methods employ the principles of class-based language modeling – the first one exploits task-specific knowledge, the second one is fully automatic and task independent. Extensive test ...

In this paper we describe the system for a fast phonetic/lexical searching in the large archives of the Czech holocaust testimonies. The developed system is the first step to a fulfillment of the MALACH project visions [1,2], at least as for an easier and faster acce...

In this paper methods are proposed which can be used to select a set of phonetically balanced sentences. The principle of the methods is presented and some experimental results are given. In the end of the paper the use of the proposed methods for the Czech read-spee...

In this paper a method of text-independent speaker recognition using discrete vector quantization is presented. The identification experiments were performed in a closed set of 599 speakers and two various types of features were tested: cepstral mean subtraction coeficients and mel-fr...

The paper reassumes our papers presented at the previous TSD workshops and concerns the Czech speech corpus which is being developed at the Department of Cybernetics, University of West Bohemia in Pilsen. It describes procedures of the corpus recording and annotation.

This paper proposes results of an application of a neural network on the problem of deciding whether a certain punctuation mark in Czech text is or is not the end of a sentence. It also discusses possibilities of using methods for relevant parameters extraction and compar...

The comparison of naturalness of various prosody models for synthetic speech is presented in this paper. Several listening tests of four methods for generating F0 are described together with their results and conclusions. The main goal is to compare the newly created data-driven pr...

This paper understands Golem as an artificial conscious being and the problem of Golem construction is seen as establishing appropriate causal relations between different causal domains - namely the causal domain of neural systems (either human brains or artificial neural networks) and ...

This paper introduces a new data-driven prosody model for the text-to-speech system ARTIC. The model is intended to be almost language-independent and to generate naturally sounding intonation with a link to semantics. It is based on text parametrisation using a new prosodic grammar...