Expert profile: Ph.D Anna Esposito

---

 Anna Esposito is Associate Professor in Computer Science in the Department of 
Psychology at the
Second University of Naples. She also teach the course 'Laboratory of 
Cybernetics' in the Department of
Physics at Salerno University since 1997. She is and has formerly been 
affiliated with the department of
Computer Science and Engineering at Wright State University, Dayton, OH, USA 
(Nov. 2000-Nov. 2002);
the International Institute for Advanced Scientific Studies (Italy) (since 
1995), and the Research Laboratory of Electronics (MIT) (since 1993 as Research 
Affiliate), where she developed her Ph.D thesis. She obtained her
Ph.D (1995) in Applied Mathematics and Computer Science at Naples University 
'Federico II'. She is author
of over 50 publications on International Journals and on Conference and 
Congress Proceedings with peer
review. 


Education
PhD in Applied Mathematics and Computer Science, Naples University , Italy . 
Title of Ph.D. thesis: Features extraction from speech: the acoustic attributes 
of vowels and stop consonants in Italian (in Italian). 

Massachusetts Institute of Technology, (MIT, Cambridge ), specialization on 
acoustic and perceptual analysis of speech, Speech Communication Group, 
Research Laboratory of Electronic (RLE). 

"Laurea" with "lode" in Computer Science earned at Salerno University , Italy . 
Title of thesis: A study on the behaviour and the learning process of 
deterministic neural networks: how to increase their storage capacity (in 
Italian). 


ACADEMICS 

Professor of "Fundamentals of Computer Science" and " Cross-modal analysis of 
speech, gestures and emotions through multimedia software tools" within the 
Laurea in Psychology at Second University of Naples, Department of Psychology, 
Caserta, since 2003; 

Professor of "Laboratory of Cybernetic" within the Laurea in Physics program at 
Salerno University , Department of Theoretical Physics, since 1997. 

Professor within the Master "Advanced Technologies for Information Processing 
and Communication" (sponsored by the Italian Government, M.U.R.S.T. D.M. n. 308 
of July 3 1995) for the courses: "Digital Signal Processing and Speech"; 
"Information Theory and Coding"; "Neural Networks" from 1995 to 2000. 

Supervisor of several PhD theses, Laurea theses and Project works, in 
collaboration with Italian and International Research Institutes, Universities, 
and Companies 

Research Activity
Emotional Features in Speech Language Results: An acoustic and perceptual 
analysis of the emotional state of parents toward their children has been 
performed and a recurrent neural network has been implemented for their 
classification.


Audio and Multimedia Applications for disabled persons. Results: A Multimedial 
Program aimed to help teachers of deaf students.

Acoustic and Perceptual Analysis of Gemination in Italian. Results:The 
perceptual and acoustic attributes of the stop consonants in geminate and 
non-geminate form have been measured.

Acoustic Analysis of Italian Speech, and more: Results: A set of acoustic 
features has been extracted from the spectrum shape analysis of stop consonants 
that allow to discriminate among the place of articulation.
The fundamental frequency, the formant frequencies, and the durations were 
measured throughout the vocalic portions of a number of phonemically symmetric 
Italian consonant-vowel-consonant utterances. Moreover, preliminary studies on 
Dysarthic speech were done.

Phonemes Segmentation, Phoneme Recognition, Speech Noise Cancellation and Noise 
Source Identification: Results: A set of experiments for classifying phonemes 
extracted from TIMIT have been performed. A set of experiments aimed to cancel 
noise from speech and to identify the noise source have been performed. A set 
of algorithms for text-independent phoneme segmentation have been proposed.


Incremental Learning. Results: A Neural Network Model for function 
approximation, based on Incremental Learning has been developed.

Neural Network (theory and applications) and Auditory Processing: Results: A 
neural network for speaker adaptation implemented. A new learning paradigm for 
neural nets has been developed and applied to implement a parallel neural adder 
and a neural network for correcting
and decoding binary linear code. A model of the auditory apparatus was 
developed and implemented.



Books Edited
[1h] G. Chollet, M. G. Di Benedetto, A. Esposito, M. Marinaro, 1999. Speech 
Processing, Recognition, and Artificial Neural Networks. Springer-Verlag 
Publisher.



Recent Pubblications

7] Shengli Fu, R. Gutierrez-Osuna, A. Esposito , Oscar N. Garcia, 2003, 
Audio/Visual Mapping with Cross-Modal Hidden Markov Models. In press on IEEE 
Transactions on Multimedia. 

[6] R. Gutierrez-Osuna, P. Kakumanu , A. Esposito , O. N. Garcia, A. Bojorquez, 
J. Castello, I. Rudomin, 2003 . Speech-driven facial animation with realistic 
dynamics. In press on IEEE Transactions on Multimedia. 

[5] E. Del Pezzo, A. Esposito , F. Giudicepietro, M. Marinaro, M. Martini, S. 
Scarpetta , 2003, Seismological discrimination of earthquakes and underwater 
explosions using neural networks . Bulletin of the Seismologic Society of 
America (BSSA), vol.93, n.1, pp. 215-223. 

[4] A. Esposito , E. C. Ezin, M. Marinaro, 2003 Mathematical Modelling of 
Passage Dynamic Function. To be published on the Proceedings of 
WILF2003,International Workshop on Fuzzy Logic and Application, Napoli, 9-11 
October 2003. 

[3] A. Esposito, P. D' Andria , 2003. " An adaptive learning algorithm for 
ECG noise and baseline drift removal" . In LNCS 2859, WIRN VIETRI 2003, B. 
Apolloni, M. Marinaro, R. Tagliaferri Eds, pp. 139-147, Sprinter-Verlag Berlin 
Heidelberg. 

[2] G. Aversano , A. Esposito, G. Chollet , 2003. " A JAVA T Interface for 
Speech Analysis and Segmentation " . In Proceedings of ISCA tutorial and 
Research Workshop on Non-linear Speech Processing (NOLISP03), 20-23 May, Le 
Croisic , France . 

[1] A. Esposito , A. Natale, S. Duncan , D. McNeill , F. Quek , 2003 . 
"Relazioni tra le pause nel parlato e le pause nei gesti: una ipotesi di 
sincronizzazione". In Proceedings of the V o Congresso Nazionale delle Sezione 
di Psicologia Sociale, Bari, Italy, 26-28 Settembre 2003.