Title | ||
---|---|---|
Synthetic speech detection using fundamental frequency variation and spectral features. |
Abstract | ||
---|---|---|
•Proposed synthetic speech detection using score fusion of CQCC, APGDF and fundamental frequency variation (FFV) features.•Best spoofing detection performance on the ASVspoof 2015 evaluation dataset with an overall EER of 0.05%.•Produced the state-of-the-art performance for ASV integrated with countermeasure framework.•Superior performance in generalization ability assessment.
|
Year | DOI | Venue |
---|---|---|
2018 | 10.1016/j.csl.2017.10.001 | Computer Speech & Language |
Keywords | Field | DocType |
All-pole group delay function (APGDF),Anti-spoofing,Constant Q cepstral coefficient (CQCC),Fundamental frequency variation (FFV),Score-level fusion,Spoofing attack | Mel-frequency cepstrum,Spoofing attack,Computer science,Voice activity detection,A priori and a posteriori,Word error rate,Speech recognition,Classifier (linguistics),Mixture model,Performance improvement | Journal |
Volume | Issue | ISSN |
48 | C | 0885-2308 |
Citations | PageRank | References |
4 | 0.39 | 33 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Monisankha Pal | 1 | 25 | 2.41 |
Dipjyoti Paul | 2 | 21 | 3.76 |
Goutam Saha | 3 | 255 | 23.17 |