Stelios Bakamidis

Speech enhancement from noise: a regenerative approach

Abstract In this paper a speech enhancement technique is proposed based on principal component analysis and a new criterion for the selection of the parsimonious number of components for noise-free signal regeneration. Both isolated phonemes and continuous speech experiments are presented. The results have been evaluated by informal listening and SNR computations, which show that the methodology has an improved performance compared to existing techniques.

ASR for emotional speech: Clarifying the issues and enhancing performance

There are multiple reasons to expect that recognising the verbal content of emotional speech will be a difficult problem, and recognition rates reported in the literature are in fact low. Including information about prosody improves recognition rate for emotions simulated by actors, but its relevance to the freer patterns of spontaneous speech is unproven. This paper shows that recognition rate for spontaneous emotionally coloured speech can be improved by using a language model based on increased representation of emotional utterances. The models are derived by adapting an already existing corpus, the British National Corpus (BNC). An emotional lexicon is used to identify emotionally coloured words, and sentences containing these words are recombined with the BNC to form a corpus with a raised proportion of emotional material. Using a language model based on that technique improves recognition rate by about 20%.

Phonemia a phoneme transcription system for speech synthesis in modern gre ek

Abstract This work gives a general presentation and classification of the various rules for text-to-phoneme transcription in Modern Greek. It is the outcome of a detailed study of Modern Greek based on more than 5000 words taken from everyday texts. We believe that this study is reasonably exhaustive, and that the rules formulated can accomodate not only texts from the Computer Science environment, but also specialized data bases referring to various subjects such as medicine, engineering etc. As this work is related to speech synthesis, the presentation has more of an engineering than a linguistic flavour. This means that the reader should not expect a strict linguistic rigour. The various rules found have led to the formulation of a computer program labelled PHONEMIA, which is the first module of a complete text-to-speech transcription project. This program has also been used in dealing with the diphones encountered in Greek.

Making assistive reading tools user friendly: a new platform for Greek dyslexic students empowered by automatic speech recognition

This work presents our effort to incorporate a state of the art speech recognition engine into a new platform for assistive reading for improving reading ability of Greek dyslexic students. This platform was developed in the framework of the Agent-DYSL, IST project, and facilitates dyslexic children in learning to read fluently. Unlike previously presented approaches, the aim of the system is not only to enable access to the reading materials within an inclusive learning system but to promote the development of reading skills by adjusting and adapting in the light of feedback to the system. The idea is to improve speech recognition performance so that gradually increase the reading capabilities of the user, gradually diminish the assistance provided, till he is able to read as a non-dyslexic reader. The evaluation results show that both learners’ reading pace and learners’ reading accuracy were increased.

Signal enhancement for continuous speech recognition

This paper presents a comparison between two parametric methods for Signal Enhancement in order to address the problem of robust Automatic Speech Recognition (ASR). An SVD-based technique (ISE) and a non-linear spectral subtraction method (NSS), have been evaluated by means of the Continuous Speech Recognition system that is used in the ERMIS project. The input signal is corrupted with coloured noise with variable signal-to-noise ratio. It was found that fine-tuning of the various parameters of the enhancement techniques is crucial for efficient optimisation of their performance. Both methods provide significant improvement of the speech recogniser performance in the presence of coloured noise, with the NSS method being slightly better.

Emotion in speech: towards an integration of linguistic, paralinguistic, and psychological analysis

If speech analysis is to detect a speakers emotional state, it needs to derive information from both linguistic information, i.e., the qualitative targets that the speaker has attained (or approximated), conforming to the rules of language; and paralinguistic information, i.e., allowed variations in the way that qualitative linguistic targets are realised. It also needs an appropriate representation of emotional states. The ERMIS project addresses the integration problem that those requirements pose. It mainly comprises a paralinguistic analysis and a robust speech recognition module. Descriptions of emotionality are derived from these modules following psychological and linguistic research that indicates the information likely to be available. We argue that progress in registering emotional states depends on establishing an overall framework of at least this level of complexity.

Robust speech recognition in the presence of noise using medical data

This paper discusses the improvement of speech recognition in the presence of noise, when a parametric method of signal enhancement is used. The speech enhancement method improves the performance of voice control MRI. This is important since errors in the presence of noise are more frequent and tend to make applications, such as spoken dialogue systems, too cumbersome to use. The input signal is corrupted with MRI noise with varying signal-to-noise ratio. A non-linear spectral subtraction method (NSS) as well as an SVD based noise reduction techniques (ISE) are used in conjunction with the Speech Recognition system of the FAST project, to quantify the impact of speech enhancement.

A novel technique for words reordering based on N-grams

This paper presents an approach for repairing word order errors in English text by reordering words in a sentence and choosing the version that maximizes the number of trigram hits according to a language model. The novelty of this method concerns the use of an efficient confusion matrix technique for reordering the words. For further reducing the number of permutations the use of unigramspsila probability is used. The comparative advantage of this method is that works with a large set of words, and avoids the laborious and costly process of collecting word order errors for creating error patterns.

Order determination and optimum harmonic reconstruction of quasi-periodic signals in noise

Abstract This article proposes a new method for determining the order of wide-band quasi-periodic signals from frequency estimates provided either by their short-time Fourier or linear prediction (LP) spectra. The method consists in the search for harmonic patterns in the signal spectrum that minimize an error sum of the estimated frequencies. This error can be thought of as an extension to the greatest common divisor of a set of spectral lines and operationally resembles to a spectral comb with teeth aperture and width varying as a function of frequency and comb order. The method performs well in noisy environments and when LP spectra are used the method is insensitive to LP model order selection. After the best harmonic structure has been estimated the original signal can be reconstructed with a minimum number of parameters.

Signal decomposition in terms of non-orthogonal sinusoidal bases

This paper presents a new decomposition for exact representation of a signal in terms of sinusoids with arbitrary frequencies. It is shown that the choice of N2 distinct frequencies can provide a set of N linearly independent sine and cosine vectors in the N-dimensional space which in turn enable the exact decomposition of any N samples signal. A dramatic improvement of the spectral resolution is achieved which does not depend on the number of available samples. The frequencies of this decomposition are not predetermined and may be selected according to the needs of the problem. Based on this decomposition, a new algorithm is presented for measuring the frequency of a sinusoid.