This paper describes data-driven modelling of all three basic prosodic features - fundamental frequency, intensity and segmental duration - in the Czech text-to-speech system ARTIC. The fundamental frequency is generated by a model based on concatenation of automatically acquired intonational pat...

We describe a statistical method for assignment of prosodic phrases and semantic accents in read speech data. The method is based on statistical evaluation of listening test data by a maximum-likelihood approach with parameters estimated by an EM algorithm. We also present linguisticall...

The present paper understands prosodic phrases as units which take part in constituting the rhythmical structure of speech. Due to very subjective and inconsistent criteria for the prosodic phrase perception there must be an objectively underlain method for prosodic phrase assignment. This&#...

We address the question of 3D face recognition and expression understanding under adverse conditions like illumination, pose, and accessories. We therefore conduct a campaign to build a 3D face database including systematic variation of poses, different types of occlusions, and a rich s...

This paper concentrates on experiments with automatic creation of queries from natural language topics, suitable for use in the Extended Boolean information retrieval system. Because of the lack and/or inadequacy of the available methods, we propose a new method, based on pairing t...

The paper presents a module for topic identification that is embedded into a complex system for acquisition and storing large volumes of text data from the Web. The module processes each of the acquired data items and assigns keywords to them from a defined topic hierarch...

Artificial neural networks have become state-of-the-art in the task of language modelling on a small corpora. While feed-forward networks are able to take into account only a fixed context length to predict the next word, recurrent neural networks (RNN) can take advantage of all&#x...

The goal of metadata extraction (MDE) is to enable technology that can take raw speech-to-text output and refine it into forms that are more useful to humans and to downstream automatic processes. Starting in 2003, a structural metadata annotation task was defined for English ...

The aim of this paper is to present an extension of the hidden vector state semantic parser. First, we describe the statistical semantic parsing and its decomposition into the semantic and the lexical model. Subsequently, we present the original hidden vector state parser. Then,&#x...

In the paper, we present a comparative study of several methods used nowadays in the field of feature and information extraction. We compared several Independent Component Analysis (ICA) algorithms together with the commonly used Principal Component Analysis (PCA) algorithm in two real-world...

In the paper we demonstrate a complex supervised learning method based on a binary decision graphs. This method is employed in construction of a silence/speech detector. Performance of the resulting silence/speech detector is compared with performance of common silence/speech detectors&...

In this paper we present a novel method for adaptation of a multi-layer perceptron neural network (MLP ANN). Nowadays, the adaptation of the ANN is usually done as an incremental retraining either of a subset or the complete set of the ANN parameters. However,�...

In the paper we give a brief introduction into sign language recognition and present a particular research task, where the access to MetaCentrum computing facilities was highly beneficial. Although the problem of signed speech recognition is currently being researched into by many resea...

In the paper we introduce the on-line captioning system developed by our teams and used by the Czech Television (CTV), the public service broadcaster in the Czech Republic. The research project is targeted at incorporation of speech technologies into the CTV environment. One of...

In the paper we present two techniques improving the recognition accuracy of multilayer perceptron neural networks (MLP ANN) by means of adopting Speaker Adaptive Training. The use of the MLP ANN, usually in combination with the TRAPS parametrization, includes applications in speech rec...

In this paper we discuss the design, acquisition and preprocessing of a Czech audio-visual speech corpus. The corpus is intended for training and testing of existing audio-visual speech recognition system. The name of the database is UWB-07-ICAVR, where ICAVR stands for Impaired Conditi...

In this paper we present a highly optimized implementation of Gaussian mixture acoustic model evaluation algorithm. Evaluation of these likelihoods is one of the most computationally intensive parts of automatics speech recognizers but it can be well-parallelized and offloaded ...

Gaussian mixture models (GMMs) are often used in various data processing and classification tasks to model a continuous probability density in a multi-dimensional space. In cases, where the dimension of the feature space is relatively high (e.g. in the automatic speech recognition (ASR)...

Traditional discriminative training methods modify Hidden Markov Model (HMM) parameters obtained via a Maximum Likelihood (ML) criterion based estimator. In this paper, anti-models are introduced instead. The anti-models are used in tandem with ML models to incorporate a discriminative information...

The main goal of this paper is to explore the methods of gender-dependent acoustic modeling that would take the possibly of imperfect function of a gender detector into consideration. Such methods will be beneficial in real-time recognition tasks (eg. real-time subtitling of meetings)&#...