| H90-1034 | no preprocessing , and with the | SDCN | and CDCN algorithms . Use of |
| H92-1055 | SDCN algorithm . Compared to the | SDCN | algorithm , the CDCN algorithm |
| H90-1034 | performance . The much simpler | SDCN | algorithm also provides considerable |
| H90-1034 | long-term averages axe used , | SDCN | is clearly not able to model |
| H90-1034 | SNR . SNR is estimated in the | SDCN | algorithm as z ( 0 ) - n ( 0 |
| H90-1034 | MMSEN is similar in concept to | SDCN | except that it operates in the |
| H90-1034 | the CRPZM with no processing , | SDCN | , and CDCN respectively . While |
| H92-1055 | Boll \ -LSB- 7 \ -RSB- ) . The | SDCN | algorithm is simple and effective |
| H92-1055 | good recognition accuracy . The | SDCN | algorithm , on the other hand |
| H90-1034 | suppression is achieved with both | SDCN | and CDCN , the CDCN algorithm |
| H90-1034 | domain . As is seen in Table 4 , | SDCN | performs slightly better than |
| H92-1055 | SNR-Dependent Cepstral Normalization ( | SDCN | ) , applies an additive correction |
| H92-1055 | computationally demanding than the | SDCN | algorithm . Compared to the SDCN |
| H90-1034 | above , the correction vector in | SDCN | , w , has the asymptotic value |
| H90-1034 | at high SNR . In interpolated | SDCN | ( ISDCN ) the dependence on SNR |
| H90-1034 | also describe an interpolated | SDCN | algorithm ( iSDCN ) which combines |
| H90-1034 | of speech ( a codebook ) , the | SDCN | can estimate the parameters of |
| H90-1034 | SNR-Dependent Cepstral Normalization ( | SDCN | ) is a simple algorithm that |
| H90-1034 | instantaneous SNR of the input . While | SDCN | is very simple , it provides |
| H90-1034 | . CDCN Algorithm Although the | SDCN | technique performs acceptably |
