Uniform and warped low delay filter-banks for speech enhancement

Löllmann, Heinrich W.; Vary, Peter

doi:10.1016/j.specom.2007.04.009

Cited by 26 publications

(26 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The quality of the processed speech is evaluated in the frequency-domain by means of the cepstral distance (CD) between the (anechoic) speech signal of the direct path sd(k) and the enhanced speechŝ(k) after delay compensation (cf., [2,9]). The segmental SIR is used as time-domain measure for the speech quality according to (cf., [4]) The set Fs contains all frame indices corresponding to frames with speech activity and C(Fs) represents its total number of elements.…”

Section: Discussionmentioning

confidence: 99%

“…Eq. (3) can be either implemented by the overlap-add method [8], or the concept of the filter-bank equalizer [9], which is considered here. This approach allows for a low algorithmic signal delay and is beneficial, if the vantages of a non-uniform (Bark-scaled) frequency resolution should be exploited, e.g., for speech enhancement in hearing aids, cf., [9].…”

Section: Low Delay Filteringmentioning

confidence: 99%

“…It is also possible to approximate the (uniform or warped) time-domain filter by an FIR or IIR filter of lower degree to further reduce signal delay and complexity. A more detailed description can be found in [9].…”

Section: Low Delay Filteringmentioning

confidence: 99%

“…This filter-bank configuration is used as the employed instrumental measures are sensitive towards filtering with non-linear phase, even if this has no audible effect, cf.,[9].…”

mentioning

confidence: 99%

See 3 more Smart Citations

A blind speech enhancement algorithm for the suppression of late reverberation and noise

Löllmann

Vary

2009

2009 IEEE International Conference on Acoustics, Speech and Signal Processing

Self Cite

View full text Add to dashboard Cite

This paper proposes a new speech enhancement algorithm for the suppression of background noise and late reverberation without a priori knowledge. A generalized spectral weighting rule based on estimations for the spectral variances of the late reverberant speech and background noise is used for speech enhancement. This allows to suppress speech distortions due to late room reflections without knowledge about the complete room impulse response. In contrast to existing methods, all needed quantities are estimated entire blindly from the reverberant and noisy speech signal. The new algorithm has also a low signal delay which is important, e.g., for speech enhancement in mobile communication devices or hearing aids.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Low Delay Filteringmentioning

confidence: 99%

Section: Low Delay Filteringmentioning

confidence: 99%

“…This filter-bank configuration is used as the employed instrumental measures are sensitive towards filtering with non-linear phase, even if this has no audible effect, cf.,[9].…”

mentioning

confidence: 99%

See 2 more Smart Citations

A blind speech enhancement algorithm for the suppression of late reverberation and noise

Löllmann

Vary

2009

2009 IEEE International Conference on Acoustics, Speech and Signal Processing

Self Cite

View full text Add to dashboard Cite

show abstract

“…The discussed filter approximations can also be applied to the allpass transformed FBE [47]. In the process, the delay elements of the analysis filter-bank and the time-domain filter are replaced by allpass filters according to Eq.…”

Section: Warped Filter Approximationmentioning

confidence: 99%