Bojan Kotnik

Bojan Kotnik

Auto-generated profile, based on publicly available information

Bojan Kotnik

Faculty of Electrical Engineering and Computer Science, University of Maribor

Research Focus & Metrics

Academic Career Length

22Years (2005-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
30 March 2005 in EURASIP Journal on Advances in Signal Processing
Verified
Journal Article
Open access
PDF file
DVDamjan Vlaj
Damjan Vlaj
BKBojan Kotnik
Bojan Kotnik
BHBogomir Horvat
Bogomir Horvat
ZKZdravko Kačič
Zdravko Kačič
This paper presents a novel computationally efficient voice activity detection (VAD) algorithm and emphasizes the importance of such algorithms in distributed speech recognition (DSR) systems. When using VAD algorithms in telecommunication systems, the required capacity of the speech transmission channel can be reduced if only the speech parts of the signal are transmitted. A similar objective can be adopted in DSR systems, where the nonspeech parameters are not sent over the transmission channel. A novel approach is proposed for VAD decisions based on mel-filter bank (MFB) outputs with the so-called Hangover criterion. Comparative tests are presented between the presented MFB VAD algorithm and three VAD algorithms used in the G.729, G.723.1, and DSR (advanced front-end) Standards. These tests were made on the Aurora 2 database, with different signal-to-noise (SNRs) ratios. In the speech recognition tests, the proposed MFB VAD outperformed all the three VAD algorithms used in the standards by relative (G.723.1 VAD), by relative (G.729 VAD), and by relative (DSR VAD) in all SNRs.

10 June 2007 in EURASIP Journal on Advances in Signal Processing
Verified
Journal Article
Open access
PDF file
BKBojan Kotnik
Bojan Kotnik
ZKZdravko Kačič
Zdravko Kačič
This paper concerns the problem of automatic speech recognition in noise-intense and adverse environments. The main goal of the proposed work is the definition, implementation, and evaluation of a novel noise robust speech signal parameterization algorithm. The proposed procedure is based on time-frequency speech signal representation using wavelet packet decomposition. A new modified soft thresholding algorithm based on time-frequency adaptive threshold determination was developed to efficiently reduce the level of additive noise in the input noisy speech signal. A two-stage Gaussian mixture model (GMM)-based classifier was developed to perform speech/nonspeech as well as voiced/unvoiced classification. The adaptive topology of the wavelet packet decomposition tree based on voiced/unvoiced detection was introduced to separately analyze voiced and unvoiced segments of the speech signal. The main feature vector consists of a combination of log-root compressed wavelet packet parameters, and autoregressive parameters. The final output feature vector is produced using a two-staged feature vector postprocessing procedure. In the experimental framework, the noisy speech databases Aurora 2 and Aurora 3 were applied together with corresponding standardized acoustical model training/testing procedures. The automatic speech recognition performance achieved using the proposed noise robust speech parameterization procedure was compared to the standardized mel-frequency cepstral coefficient (MFCC) feature extraction procedures ETSI ES 201 108 and ETSI ES 202 050.

2 May 2006
Verified
Conference Paper
Open access
PDF file
BKBojan Kotnik
Bojan Kotnik
HHHarald Höge
Harald Höge
ZKZdravko Kacic
Zdravko Kacic
Robust fundamental frequency estimation in adverse conditions is important in various speech processing applications. In this paper a new pitch detection algorithm (PDA) based on the autocorrelation of the Hilbert envelope of the LP residual [1] is compared to another well established algorithm from Goncharoff [2]. A set of evaluation criteria is collected on which the two PDA algorithms are compared. In order to evaluate the algorithms in adverse conditions a suited reference database was constructed. This reference database consists of parts of the SPEECON speech database [3] where recordings of 60 speakers were selected and manually pitch marked. The recordings cover several adverse conditions as noise in the car cabin and reverberations of office rooms. The evaluation highlights the good performance of the new algorithm in comparison but shows, that low SNR conditions and strong reverberation are still a demanding challenge for future pitch detection algorithms.