1.  Wavelet Analysis and Synthesis
 
The use of wavelets for pseudo random signals and signals that change over time is now common. This project aims at developing software for the synthesis of a signal whether speech or music using a set of wavelet basis functions. It also aims at looking into the coding properties of wavelets. Requirements are subject knowledge and software tools available.
 
 2. Speech synthesis using Harmonic + Noise Model
 
This project aims at developing software for speech synthesis using a newer model of speech. The student will be expected to learn the model, use speech data for analysis, development of the model parameters and eventual synthesis from the model parameters. Requirements are subject knowledge, existing speech corpora and software tools.
This area is of high topical interest because of the move towards direct speech machine - human interface
 
3. Automatic Segmentation of Speech using Neural Network
 
This is an ongoing project aimed at building annotated Maltese speech corpora. The first project used the American TIMIT corpus to build a phoneme based neural net. This was adapted in a second project to test Maltese sentences and adapt the American phoneme set to the Maltese phoneme set. The aim of this project is now to use the existing neural net and appropriate software to overcome the present limitations, to automatically segment existing Maltese speech corpora into its phonemic content, to be able to use this annotated corpus to train a Maltese based speech recognition system.
 
4. Speech Recognition in Noisy Background
 
This project is a sequel to last year's project. It aims at looking into appropriate robust features that are immune to background noise ,such as in an office, while talking. The student will have to familiarise himself/herself with acoustics of speech and various speech parameters. Preliminary testing need not necessarily use neural nets or HMM.

5. Speech pedagogical tools
 
This project is suitable both for engineering and IT students. It aims at producing a friendly user interface. There are various tools that can be developed, and the student will be concerned with one of them. Among others, a GUI showing time waveforms, acquisition of parameters on the sliding time waveform. Parameters including FFT, spectral envelope, various acoustic features.
Another is a GUI for various neural net types, aimed at showing the weights varying during training, effect of hidden layer node number etc.

6. Spoken Dialogue System for Lotto
 
This aims at developing a natural language model interlinked with a speech recognition system that is capable of recognising the bets on the various games such as lotto, superfive etc., the numbers being betted, and the price or type. The machine has to perform a suitable dialogue. This year there can be various developments including the natural language definitions, (suitable for IT students), a start on suitable recording of speech to extract training material for the recogniser, etc.
Requirements are a course on linguistics or natural language understanding and the student would have to learn on speech recognition systems. Could be a shared project between an IT and an engineering student in collaboration to develop the system

7. Acoustic Modelling

The continuation of a project started this year on acoustic modelling. A basic program model has been developed and it is intended to continue the development. Acoustic modelling implies the calculation of parameters based on different models and there use to obtain the response at a receiver based in a room. In this way room acoustics can be modelled prior to building. Particularly in this area there can be subsets of work including: (i) A formal analysis of the acoustic properties of Maltese building materials. This could also require input from the Architecture Department, with whom there can be collaboration on this. (ii) The building of a head auralisation model to simulate the binaural model of hearing as opposed to the omnidirectional model of reception.