Restoration of audio documents by means of extended Kalman filter

Authors:
Sergio Canazza;Giovanni De Poli;Gian Antonio Mian
Affiliations:
Department of Historical and Documentary Sciences, University of Udine, Udine, Italy;Department of Information Engineering, University of Padova, Padova, Italy;Department of Information Engineering, University of Padova, Padova, Italy
Venue:
IEEE Transactions on Audio, Speech, and Language Processing
Year:
2010

Citing 9
Cited 1

Steady-state and parameter tracking properties of self-tuning minimum variance regulators

Automatica (Journal of IFAC)
Fundamentals of speech recognition

Fundamentals of speech recognition
Identification of Time-Varying Processes

Identification of Time-Varying Processes
Linear Prediction of Speech

Linear Prediction of Speech
Adaptive scheme for elimination of broadband noise and impulsivedisturbances from AR and ARMA signals

IEEE Transactions on Signal Processing
An efficient model-based multirate method for reconstruction of audio signals across long gaps

IEEE Transactions on Audio, Speech, and Language Processing
Analysis and Comparison of Multichannel Noise Reduction Methods in a Common Framework

IEEE Transactions on Audio, Speech, and Language Processing
Speech enhancement using a masking threshold constrained Kalman filter and its heuristic implementations

IEEE Transactions on Audio, Speech, and Language Processing
On causal algorithms for speech enhancement

IEEE Transactions on Audio, Speech, and Language Processing

The restoration of low-quality audio recordings based on non-negative matrix factorization and perceptual assessment by means of the ebu mushra test method

Proceedings of the second workshop on eHeritage and digital art preservation

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present some results on audio restoration obtained with an algorithm that solves the problems of broadband noise filtering, signal parameters tracking, and impulsive noise removal by using the Extended Kalman Filter (EKF) theory. We show that, to achieve maximum performance, it is essential to optimize the EKF implementation. To this purpose, having to cope with the nonstationarity of the audio signal, we use two properly combined EKF filters (forward and backward), and introduce a bootstrapping procedure for model tracking. The careful combination of the proposed techniques and an accurate choice of some critical parameters, allows to improve the performance of the EKF algorithm. The presented procedure is validated by listening tests.