Researcher
  • Phone: +39 0461314564
  • FBK Povo
Short bio

Piergiorgio Svaizer received his degree in Electronic Engineering from the University of Padova in 1989. In 1991 he joined ITC-IRST (now FBK-IRST) where he has been working in the Speech Recognition Group and then in the SHINE research unit. His current research interests include digital signal processing, microphone arrays, automatic speech recognition in noisy and reverberant environments.

Research interests
digital signal processing Microphone arrays Acoustic Scene Analysis Automatic speech recognition in noisy and reverberant environments
Publications
  1. M. Matassoni; G. A. Mian; M. Omologo; A. Santarelli; P. Svaizer,
    Some experiments on the use of one-channel noise reduction techniques with the Italian SpeechDatCar database,
    IEEE Workshop on Automatic Speech Recognition and Understanding, 2001,
    IEEE,
    2001
    , pp. 139-
    142
    , (Workshop on Automatic Speech Recognition and Understanding,
    Madonna di Campiglio, Italy,
    da 9/12/2001 a 13/12//2001)
  2. Luca Cristoforetti; Marco Matassoni; Maurizio Omologo; Piergiorgio Svaizer; Enrico Zovato,
    Annotation of a multichannel noisy speech corpus,
    International Conference on Language Resources and Evaluation (LREC 2000),
    2000
    , pp. 1547-
    1550
    , (International Conference on Language Resources and Evaluation (LREC 2000),
    Athens, Greece,
    30/05/2000 - 02/06/2000)
  3. E. Zovato; Piergiorgio Svaizer; Maurizio Omologo; Marco Matassoni,
    Un sistema di riconoscimento vocale automatico operante in automobile,
    Proceedings of the 28° Convegno Nazionale Associazione Italiana di Acustica [AIA],
    2000
    , pp. 271-
    274
  4. Marco Matassoni; Maurizio Omologo; Diego Giuliani; Piergiorgio Svaizer,
    Training of HMM with Contaminated Speech Material for Hands-Free Speech Recognition,
    Challenging scenario is addressed in which a hands-free speech recognizer operates in a noisy office environment with model adaptation functionalities. The use of a single far microphone as well as that of a microphone array input are investigated. beside the benefits due to the application of microphone array processing, system robustness is improved by training hidden Markov models with a contaminated version of a clean corpus. This artificial corpus is produced by exploiting information extracted from "real world" acoustic scenarios. The resulting models are then used as starting point for unsupervised incremental adaptation. Experimental results of connected digit recognition in a real noisy environment show the advantages provided by the joint use of microphone array processing, HMM training on contaminated speech, and incremental adaptation, as well as their respective contribution to the overall improvement of performance, which started from approximately 30% word recognition rate using the baseline system and achieved 99% using the best system configuration,
    2000
  5. Marco Matassoni; Maurizio Omologo; Piergiorgio Svaizer,
    A Baseline System for Hands-Free Speech Recognition in car environment,
    Robust hands-free interaction is required for a wide diffusion of automatic speech recognition in the car environment. Under the European projects SpeechDatCar and VODIS~II we collected an Italian database of in-car speech consisting of 600 sessions recorded under various noise and driving conditions. In this paper we describe some recognition experiments on two tasks devised on a portion of this database: connected digit sequences and isolated command words. Performance obtained with HMMs trained on real data collected in the car is compared with that achievable with a speech contamination approach, which aims at reproducing realistic data on the basis of a clean speech corpus. Recognition performance is also analyzed as a function of the different noisy conditions and of the consequent SNR at the far microphones. Finally, the effect of HMM adaptation is investigated in order to tune the recognizer on the conditions of the various sessions,
    2000
  6. Diego Giuliani; Marco Matassoni; Maurizio Omologo; Piergiorgio Svaizer,
    Training of HMM with filtered speech material for hands-free recognition,
    IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP),
    1999
    , pp. 449-
    452
    , (IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP),
    Phoenix, US,
    15/03/1999 - 19/03/1999)
  7. Diego Giuliani; Marco Matassoni; Maurizio Omologo; Piergiorgio Svaizer,
    Use of Filtered Clean Speech for Robust HMM Training,
    Workshop on Robust Methods for Speech Recognition in Adverse Conditions,
    1999
    , pp. 99-
    102
    , (Workshop on Robust Methods for Speech Recognition in Adverse Conditions,
    Tampere, Finland,
    25/05/1999 - 26/05/1999)
  8. Diego Giuliani; Marco Matassoni; Maurizio Omologo; Piergiorgio Svaizer,
    Robust HMM training and adaptation in hands-free speech recognition,
    IEEE Workshop on Automatic Speech Recognition and Understanding,
    1999
    , pp. 51-
    54
    , (IEEE Workshop on Automatic Speech Recognition and Understanding,
    Keystone, US,
    12/1999)
  9. Marco Matassoni; Maurizio Omologo; Luca Cristoforetti; Diego Giuliani; Piergiorgio Svaizer; Edmondo Trentin; E. Zovato,
    Some results on the development of a hands-free speech recognizer for car-environment,
    IEEE Workshop on Automatic Speech Recognition and Understanding,
    1999
    , (IEEE Workshop on Automatic Speech Recognition and Understanding,
    Keystone, US,
    12/1999)
  10. Marco Matassoni; Maurizio Omologo; Piergiorgio Svaizer; Diego Giuliani,
    Filtering Clean Speech for Training a HMM-based Hands-Free Recognizer,
    Hands-free continuous speech recognition represents a challenging scenario. In the last years, many experimental activities were devoted to investigate the use of both single microphones and microphone arrays as input to a hands-free speech recognizer. In this work, the use of a Hidden Markov Model (HMM) recognizer based on a microphone array input is considered. HMM training is accomplished by using a corpus of filtered speech material previously preprocessed in order to reproduce realistic reverberation and noise effects,
    1999

Pages