### Table 4: Performances of the online unsupervised adaptation of the system along with the missed adaptation (MAd) and false adaptation (FAd) errors, as a function of the MAP adaptation weight a6 and the adaptation threshold a78 .

2004

"... In PAGE 4: ... However the system is much less sensitive to this threshold than to the hard decision threshold a78 , this is an advantage for practical use. The Detection Error Tradeoff (DET) curves for both the on- line supervised and unsupervised systems are shown along with the DET curve of the baseline system (cf Table4 ). The unsuper- vised online adaptation of the system seems to provide perfor- mance balanced along the DET curve between the baseline and the supervised adaptation setup, except for low miss detections where it gets closer of the baseline system: the unsupervised adaptation did not improve performances for the target speak- ers with the lowest detection scores.... ..."

Cited by 1

### Table 1: Good degree sequences of rate one-half for the AWGNC and with maximal left degrees dl = 4; 5; 6; 7; 8; 9; 10; 11 and 12. For each sequence the threshold value , the corresponding ( Eb N0 ) in dB and p = Q(1= ), i.e., the input bit error probability of a hard-decision decoder are given. Also listed is 2, the maximal stable value of 2 for the given 0(1) and for = .

1999

"... In PAGE 22: ... The result of our search for the AWGNC is summarized in Tables 1 and 2. Table1 contains those sequences with dl = 4; ; 12, whereas Table 2 contains sequences with dl = 15; 20; 30 and 50. In each table columns correspond to one particular degree sequence.... ..."

Cited by 24

### Table 2: Good degree sequences of rate one-half for the AWGNC and with maximal left degrees dl = 15; 20; 30 and 50. For each sequence the threshold value , the corresponding ( Eb N0 ) in dB and p = Q(1= ), i.e., the input bit error probability of a hard-decision decoder are given. Also listed is 2, the maximal stable value of 2 for the given 0(1) and for = .

1999

"... In PAGE 22: ... The result of our search for the AWGNC is summarized in Tables 1 and 2. Table 1 contains those sequences with dl = 4; ; 12, whereas Table2 contains sequences with dl = 15; 20; 30 and 50. In each table columns correspond to one particular degree sequence.... ..."

Cited by 24

### Table 2: Good degree sequences of rate one-half for the AWGNC and with maximal left degrees dl = 15; 20; 30 and 50. For each sequence the threshold value , the corresponding ( Eb N0 ) in dB and p = Q(1= ), i.e., the input bit error probability of a hard-decision decoder are given. Also listed is 2, the maximal stable value of 2 for the given 0(1) and for = .

1999

"... In PAGE 14: ... The result of our search for the AWGNC is summarized in Tables 1 and 2. Table 1 contains those sequences with dl = 4; ; 12, whereas Table2 contains sequences with dl = 15; 20; 30 and 50. In each table columns correspond to one particular degree sequence.... ..."

Cited by 24

### Table 1: Good degree sequences of rate one-half for the AWGNC and with maximal left degrees dl = 4; 5; 6; 7; 8; 9; 10; 11 and 12. For each sequence the threshold value , the corresponding ( Eb N0 ) in dB and p = Q(1= ), i.e., the input bit error probability of a hard-decision decoder are given. Also listed is 2, the maximal stable value of 2 for the given 0(1) and for = .

1999

"... In PAGE 14: ... The result of our search for the AWGNC is summarized in Tables 1 and 2. Table1 contains those sequences with dl = 4; ; 12, whereas Table 2 contains sequences with dl = 15; 20; 30 and 50. In each table columns correspond to one particular degree sequence.... ..."

Cited by 24

### Table 2: Good degree sequences of rate one-half for the AWGNC and with maximal left degrees dl = 15; 20; 30 and 50. For each sequence the threshold value , the corresponding ( Eb N0 ) in dB and p = Q(1= ), i.e., the input bit error probability of a hard-decision decoder are given. Also listed is 2, the maximal stable value of 2 for the given 0(1) and for = .

1999

"... In PAGE 14: ... The result of our search for the AWGNC is summarized in Tables 1 and 2. Table 1 contains those sequences with dl = 4; ; 12, whereas Table2 contains sequences with dl = 15; 20; 30 and 50. In each table columns correspond to one particular degree sequence.... ..."

Cited by 24

### Table 1: Good degree sequences of rate one-half for the AWGNC and with maximal left degrees dl = 4; 5; 6; 7; 8; 9; 10; 11 and 12. For each sequence the threshold value , the corresponding ( Eb N0 ) in dB and p = Q(1= ), i.e., the input bit error probability of a hard-decision decoder are given. Also listed is 2, the maximal stable value of 2 for the given 0(1) and for = .

1999

"... In PAGE 14: ... The result of our search for the AWGNC is summarized in Tables 1 and 2. Table1 contains those sequences with dl = 4; ; 12, whereas Table 2 contains sequences with dl = 15; 20; 30 and 50. In each table columns correspond to one particular degree sequence.... ..."

Cited by 24

### Table 3: Hard and Soft Thresholding Standard Optimal

1998

"... In PAGE 15: ... Figure 5 shows estimates obtained using the two methods. Table3 compares hard and soft thresholding. Hard thresholding gives a lower MSE and is less critical to the number of levels that are thresholded.... ..."

Cited by 23

### Table 3: Hard and Soft Thresholding Standard Optimal

1998

"... In PAGE 5: ... Figure 4 shows estimates obtained using the two methods. Table3 compares hard and soft thresholding. Hard thresholding gives a lower MSE and is less critical to the number of levels that are thresholded.... In PAGE 9: ... Thresholding using di erent pre lters Table 2. Univariate and Bivariate thresholding Table3 . Hard and Soft Thresholding Table 4.... ..."

Cited by 23

### Table 1: Hard and soft decision OAAE values

"... In PAGE 3: ... There are a total of 31 AAE values and an average of these values result in an overall average absolute error (OAAE). Table1 depicts these OAAE values for each of the six features. Three different VQ clas- sifier systems are attempted.... ..."