Robust Feature Extraction Using Spectral Peaks of the Filtered Higher Lag Autocorrelation Sequence of the Speech Signal

Farahani, Gholamreza; Ahadi, Seyed Mohammad; Homayounr, M.M.; Kashi, Amir Abbas

doi:10.1109/isspit.2006.270925

Cited by 6 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With this differentiation, the flat parts of the spectrum were almost removed and each spectral peak was split into two, one positive and one negative. The differential power spectrum of the noisy signal was defined as where P and Q are the orders of the difference equation, a 1 are real-valued coefficients and K is the length of FFT (on the positive frequency side) [16]. The differentiation mentioned in equation (8) can be carried out in several ways, as discussed in reference [16].…”

Section: Differentiation Of Autocorrelation Sequence (Das)mentioning

confidence: 99%

“…The differential power spectrum of the noisy signal was defined as where P and Q are the orders of the difference equation, a 1 are real-valued coefficients and K is the length of FFT (on the positive frequency side) [16]. The differentiation mentioned in equation (8) can be carried out in several ways, as discussed in reference [16]. The simple difference had shown the best results and therefore was used in reference [11], i.e.…”

Section: Differentiation Of Autocorrelation Sequence (Das)mentioning

confidence: 99%

“…Also the duration for lower lag elimination in the AMFCC method was set to 2.5 ms (20 samples in 8 kHz sampling rate for Aurora 2 task) similar to reference [10]. The same duration was also used for SPFH implementation [16]. In order to estimate the noise autocorrelation sequence, in all our experiments, we have used 20 initial frames of each utterance, considering them as nonspeech sections.…”

Section: Parameter Settingsmentioning

confidence: 99%

See 2 more Smart Citations

Robust Feature Extraction Using Autocorrelation Domain for Noisy Speech Recognition

Farahani¹

2017

SIPIJ

View full text Add to dashboard Cite

show abstract

Section: Differentiation Of Autocorrelation Sequence (Das)mentioning

confidence: 99%

Section: Differentiation Of Autocorrelation Sequence (Das)mentioning

confidence: 99%

Section: Parameter Settingsmentioning

confidence: 99%

See 1 more Smart Citation

Robust Feature Extraction Using Autocorrelation Domain for Noisy Speech Recognition

Farahani¹

2017

SIPIJ

View full text Add to dashboard Cite

show abstract

“…9. The resultant feature set was called spectral peaks of filtered higher-lag autocorrelation sequence (SPFH) (Farahani et al, 2006b). …”

Section: Spectral Peaks Of Filtered Higher-lag Autocorrelation Sequenmentioning

confidence: 99%

Autocorrelation-based Methods for Noise-Robust Speech Recognition

Farahani¹,

Ahadi²,

Homayounpour³

2007

Robust Speech Recognition and Understanding

View full text Add to dashboard Cite

“…A great deal of efforts has been devoted to such techniques as RASTA filtering, cepstral mean normalization (CMN), use of dynamic spectral features, short-time modified coherence (SMC), one-sided autocorrelation LPC (OSALPC) and relative autocorrelation sequence (RAS) [5].…”

Section: Introductionmentioning

confidence: 99%

A Neural Network based local SNR estimation for estimating spectral masks

Hadjahmadi

Homayounpour

Ahadi

2008

2008 International Symposium on Telecommunications

View full text Add to dashboard Cite

In this work, we present a new mask estimation technique that uses a Neural Network classifier to determine the reliability of spectrographic elements. In addition some different kinds of features used for classification were compared that make no assumptions about the corrupting noise signal, but rather exploit spectrographic characteristics of the speech signal. The performance of the proposed method is experimentally evaluated in text independent speaker recognition task using the Gaussian Mixture Model (GMM) under various noise conditions. The speaker recognition results were achieved using the TFarsdat corpus. Noisy speech is simulated by adding noise sources taken from the NOISEX-92 database. Experimental results obtained show that the new neural network based mask estimation method is effective for speaker recognition under noisy conditions.

show abstract

Robust Feature Extraction Using Spectral Peaks of the Filtered Higher Lag Autocorrelation Sequence of the Speech Signal

Cited by 6 publications

References 8 publications

Robust Feature Extraction Using Autocorrelation Domain for Noisy Speech Recognition

Robust Feature Extraction Using Autocorrelation Domain for Noisy Speech Recognition

Autocorrelation-based Methods for Noise-Robust Speech Recognition

A Neural Network based local SNR estimation for estimating spectral masks

Contact Info

Product

Resources

About