Developing a Child Friendly Text-to-Speech System

This paper discusses the implementation details of a child friendly, good quality, English text-to-speech (TTS) system that is phoneme-based, concatenative, easy to set up and use with little memory. Direct waveform concatenation and linear prediction coding (LPC) are used. Most existing TTS systems are unit-selection based, which use standard speech databases available in neutral adult voices.Here reduced memory is achieved by the concatenation of phonemes and by replacing phonetic wave files with their LPC coefficients. Linguistic analysis was used to reduce the algorithmic complexity instead of signal processing techniques. Sufficient degree of customization and generalization catering to the needs of the child user had been included through the provision for vocabulary and voice selection to suit the requisites of the child. Prosody had also been incorporated. This inexpensive TTS systemwas implemented inMATLAB, with the synthesis presented by means of a graphical user interface (GUI), thus making it child friendly. This can be used not only as an interesting language learning aid for the normal child but it also serves as a speech aid to the vocally disabled child. The quality of the synthesized speech was evaluated using the mean opinion score (MOS).

Improving Multi-class Text Classification with Naive Bayes

There are numerous text documents available in electronic form. More and more are becoming available every day. Such documents represent a massive amount of information that is easily accessible. Seeking value in this huge collection requires organization; much of the work of organizing documents can be automated through text classification. The accuracy and our understanding of such systems greatly influences their usefulness. In this paper, we seek 1) to advance the understanding of commonly used text classification techniques, and 2) through that understanding, improve the tools that are available for text classification. We begin by clarifying the assumptions made in the derivation of Naive Bayes, noting basic properties and proposing ways for its extension and improvement. Next, we investigate the quality of Naive Bayes parameter estimates and their impact on classification. Our analysis leads to a theorem which gives an explanation for the improvements that can be found in multiclass classification with Naive Bayes using Error-Correcting Output Codes. We use experimental evidence on two commonly-used data sets to exhibit an application of the theorem. Finally, we show fundamental flaws in a commonly-used feature selection algorithm and develop a statistics-based framework for text feature selection. Greater understanding of Naive Bayes and the properties of text allows us to make better use of it in text classification.

An Introduction to Behaviour Analysis

Behaviour Analysis is a distinct philospophy of science. Individuals new to the approach often find difficulty in understanding the basic principles involved. This presentation, aimed at Final Year undergraduates, is designed to provide an introduction to the principles of operant conditioning (e.g., reinforcement, punishment, and extinction), making clear that these words describe functional, rather than structural, relations.

TU Graz: Course: 707.000 Web Science and Web Technology: Lecture 4: Social Network Analysis

What are fundamental entities in social networks and what information is contained in social graphs? We will discuss some selected concepts in social network analysis, such as one- and two mode networks, prestige and centrality, and cliques, clans and clubs. Readings: Web tool predicts election results and stock prices, J. Palmer, New Scientist, 07 February (2008) [Protected Access] Optional: Social Network Analysis, Methods and Applications, S. Wasserman and K. Faust (1994)

TU Graz: Course: 707.000 Web Science and Web Technology: Lecture 7: Link Analysis and Search

What are ways of searching in graphs? In this class, we will discuss basics of link analysis, including Google's PageRank algorithm as an example. Readings: The PageRank Citation Ranking: Bringing Order to the Web, L. Page and S. Brin and R. Motwani and T. Winograd (1998) Stanford Tecnical Report

Decision analysis exercises

Exercises and solutions in LaTex

Simple Analysis of Network Datasets

Linux commands that are generally useful for analyzing data; it is very easy to reduce phenomena such as links, nodes, URLs or downloads, to multiply repeating identifiers and then sorting and counting appearances.

El gènere del text de divulgació científica: una proposta d’aplicació a l’aula

L’objectiu d’aquest estudi és presentar una proposta de lectura i producció textual, el Text de Divulgació Científica, tot fent servir una seqüència didàctica, per tal de preparar l’alumne per a la lectura i anàlisi de l’estructura textual que presenta aquest gènere discursiu, així com despertar-li l’interès per la recerca. D’aquesta manera, l’alumne comptarà amb instruments per a la producció escrita del gènere discursiu del text de divulgació científica. La proposta es basa en els treballs d’autors de lingüística textual i anàlisi del discurs.

Ecological analysis of periphytic diatoms in Mediterranean coastal wetlands (Empordà wetlands, NE Spain)

S'han estudiat els efectes dels factors ambientals sobre el perífiton dels sistemes lenític fluctuants del aiguamolls de l'Empordà. L'estudi s'ha realitzat als tres nivells d'integració: nivell d'ecosistema considerant el rol del perífiton envers els altres productors primaris; a nivell de comunitat, estudiant la composició específica de les diatomees i a nivell de població estudiant la plasticitat fenotípica d'una espècie de diatomea (Nitzschia frustulum). A nivell d'ecosistema s'observa que els factors que afavoreixen el predomini dels diferents tipus de productors primaris (perífiton, fitoplàncton i macròfits) són la renovació i el grau d'eutròfia de l'aigua. A nivell de comunitat els factors determinants en la composició i distribució de les espècies de diatomees són els gradients confinament-inundació així com la productivitat del sistema. En funció d'aquest factors s'han establert 5 associacions de diatomees. A nivell de població es demostra que tant la salinitat, com la relació N : P a l'aigua com el moviment de l'aigua afecten la morfologia i ultraestructura de la valva de N. frustulum. De forma interessant s'observa que la salinitat, considerada com a factor individual, afecta N. frustulum a nivell poblacional provocant-li modificacions en la morfologia de la valva, per en canvi, no afecta a nivell de comunitat, ja que totes les espècies de diatomees presents en ambients de salinitat fluctuant són eurihalines.