The use of sub-band cepstrum in speaker verification

Sivakumaran, P.; Ariyaeeinia, Aladdin M.

doi:10.1109/icassp.2000.859149

Cited by 3 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to tackle this problem a modified version of the Baum-Welch re-estimation procedure is used in this work [19]. In this approach, given the training utterance O l , the probabilities ) , …”

Section: Resultsmentioning

confidence: 99%

“…The main focus of this section is the spectral information represented by the sub-band cepstral parameters in relation to that contained in the full-band cepstral parameters [18]. Suppose that a S subband system is formed by equally distributing the log spectral parameters ( )…”

Section: Sub-band Cepstrummentioning

confidence: 99%

“…Remark I: It should be noted that simple linear relationships, such as (19) and (20), would be possible between sub-band and full-band cepstral parameters only if the log-spectral magnitudes are equally divided into non-overlapping sub-bands. However, the general point in the above discussions is that the overall spectral variation measured by a lower order DCT basis function in a sub-band is comparable to that of a higher DCT basis function of the full-band in the same spectral region.…”

Section: Sub-band Cepstrummentioning

confidence: 99%

“…A logical way to tackle this problem is to split the entire frequency domain into a number of sub-regions and to use the spectral information contained in each of these regions to generate independent feature vectors. This technique is commonly known as the sub-band analysis and has been studied in the context of both speech and speaker recognition [3][5] [6][11] [17][18][19][20]. The main attraction of this approach is that it provides the possibility of selectively de-emphasising the frequency spectral regions that are affected by anomalies.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Sub-band based text-dependent speaker verification

Sivakumaran¹,

Ariyaeeinia²,

Loomes³

2003

Speech Communication

Self Cite

View full text Add to dashboard Cite

This paper addresses various issues involved in sub-band based text-dependent speaker verification.The first part of the discussions is concerned with the classification methods. An important issue addressed in this part is the determination of a set of weights which emphasises the sub-bands that are specific to the target speaker while de-emphasising or removing the contaminated ones. In particular, techniques for determining these weights dynamically according to the level of contamination in the sub-bands are described. Furthermore, the effectiveness of these methods is experimentally analysed through a set of comparative studies. The second part of the discussions focuses on the feature extraction process. Analytically, it is shown that for a sub-band system of S bands, the cepstral coefficients with the quefrency of p have a strong linear relationship to the (S×p) th full-band cepstral parameter.With the aid of a set of experimental results, it is demonstrated that this means the conventional classification methods adapted to work with sub-band cepstral parameters may not be able to capture all the useful spectral information contained in the full-band cepstral parameters. In order to tackle this problem, two methods are described and their relative effectiveness is experimentally examined.The experimental investigations also include an examination of speaker discrimination abilities of different sub-bands and an analysis of different possible recombination levels.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Sub-band Cepstrummentioning

confidence: 99%

Section: Sub-band Cepstrummentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Sub-band based text-dependent speaker verification

Sivakumaran¹,

Ariyaeeinia²,

Loomes³

2003

Speech Communication

Self Cite

View full text Add to dashboard Cite

show abstract

“…Text dependent: In text dependent, verification of the speaker's identity is based on his/her speaking of one or more specific phrases such as: password, card number and PIN number etc. [15].…”

Section: Classification Of Speaker Recognitionmentioning

confidence: 99%

Adaptive Speech Analysis Techniques for Biometrics Applications

Mozaffar¹

2021

Preprint

View full text Add to dashboard Cite

Using biometrics for personal identity authentication is becoming more reliable and accurate than current methods because it links the event to a particular individual (e.g. a password may be used by someone other that the authorized user), is convenient (nothing to carry or remember), accurate (provides positive authentication) and is socially acceptable. Adaptive frequency sub-band or time frame recombination approaches are introduced in this thesis.

show abstract

A new feature extraction based the reliability of speech in speaker recognition

Yang

Canwei

6th International Conference on Signal Processing, 2002.

View full text Add to dashboard Cite

The use of sub-band cepstrum in speaker verification

Cited by 3 publications

References 9 publications

Sub-band based text-dependent speaker verification

Sub-band based text-dependent speaker verification

Adaptive Speech Analysis Techniques for Biometrics Applications

A new feature extraction based the reliability of speech in speaker recognition

Contact Info

Product

Resources

About