Effect of multi-condition training and speech enhancement methods on spoofing detection
The DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
Many researchers have demonstrated the good performance of spoofing detection systems under clean training and testing conditions. However, it is well known that the performance of speaker and speech recognition systems significantly degrades in noisy conditions. Therefore, it is of great interest to investigate the effect of noise on the performance of spoofing detection systems. In this paper, we investigate a multi-conditional training method where spoofing detection models are trained with a mix of clean and noisy data. In addition, we study the effect of different noise types as well as speech enhancement methods on a state-of-the-art spoofing detection system based on the dynamic linear frequency cepstral coefficients (LFCC) feature and a Gaussian mixture model maximum-likelihood (GMM-ML) classifier. In the experiment part we consider three additive noise types, Cantine, Babble and white Gaussian at different signal-to-noise ratios, and two mainstream speech enhancement methods, Wiener filtering and minimum mean-square error. The experimental results show that enhancement methods are not suitable for the spoofing detection task, as the spoofing detection accuracy will be reduced after speech enhancement. Multi-conditional training, however, shows potential at reducing error rates for spoofing detection.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
“Librispeech: an ASR corpus based on public domain audio books,”
Panayotov, V., Chen, G., Povey, D., & Khudanpur, S. (2015) · 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2015.7178964 · Cited by 6,072
Front-End Factor Analysis for Speaker Verification
Dehak, N., Kenny, P. J., Dehak, R., Dumouchel, P., & Ouellet, P. (2011) · IEEE Transactions on Audio, Speech, and Language Processing · doi:10.1109/tasl.2010.2064307 · Cited by 3,549
ASVspoof 2015: the first automatic speaker verification spoofing and countermeasures challenge
Wu, Z., Kinnunen, T., Evans, N., Yamagishi, J., Hanilçi, C., Sahidullah, M., & Sizov, A. (2015) · Interspeech 2015 · doi:10.21437/interspeech.2015-462 · Cited by 484 · cited 3×
“Experiments are performed on the ASVspoof 2015 database [18] which consists of genuine human speech and ten”
BackgroundMethodA comparison of features for synthetic speech detection
Sahidullah, M., Kinnunen, T., & Hanilçi, C. (2015) · Interspeech 2015 · doi:10.21437/interspeech.2015-472 · Cited by 382 · cited 5×
“The same as in paper [17], human and spoof GMM models are trained using their corresponding labeled training data using an expectation maximization algorithm.”
BackgroundMethodInfluentialSpoofing detection with DNN and one-class SVM for the ASVspoof 2015 challenge
Villalba, J., Miguel, A., Ortega, A., & Lleida, E. (2015) · Interspeech 2015 · doi:10.21437/interspeech.2015-468 · Cited by 72 · cited 2×
“Those techniques can be divided into different domains namely feature space: Mel-frequency cepstral coefficients (MFCC) [10], linear frequency cepstral coefficients (LFCC) [12], Group Delay (GD) feature [13], relative phase shift (RPS) [14], spectral centroid magnitude coefficients (SCMC) [15], deep neural network (DN…”
Background
Cited by
ASVspoof: The Automatic Speaker Verification Spoofing and Countermeasures Challenge
Wu, Z., Yamagishi, J., Kinnunen, T., Hanilci, C., Sahidullah, M., Sizov, A., Evans, N., Todisco, M., & Delgado, H. (2017) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2017.2671435 · Cited by 229
FoR: A Dataset for Synthetic Speech Detection
Reimao, R., & Tzerpos, V. (2019) · 2019 International Conference on Speech Technology and Human-Computer Dialogue (SpeD) · doi:10.1109/sped.2019.8906599 · Cited by 193
DNN Filter Bank Cepstral Coefficients for Spoofing Detection
Yu, H., Tan, Z.-H., Zhang, Y., Ma, Z., & Guo, J. (2017) · IEEE Access · doi:10.1109/access.2017.2687041 · Cited by 67
Self-citationImpact of Bandwidth and Channel Variation on Presentation Attack Detection for Speaker Verification
Delgado, H., Todisco, M., Evans, N., Sahidullah, M., Liu, W. M., Alegre, F., Kinnunen, T., & Fauve, B. (2017) · 2017 International Conference of the Biometrics Special Interest Group (BIOSIG) · doi:10.23919/biosig.2017.8053510 · Cited by 8
“Some recent work has explored the impact of additive noise on reliability [9], [10] and the benefit of speech enhancement and multi-condition training as a means of improving robustness [9], [11].”
Background
Metadata
- Authors
- Yu, H., Sarkar, A., Thomsen, D. A. L., Tan, Z.-H., Ma, Z., & Guo, J.
- Venue
- 2016 First International Workshop on Sensing, Processing and Learning for Intelligent Machines (SPLINE)
- Publisher
- IEEE
- Published
- 2016-07
- Pages
- 1-5
- OpenAlex
- W2503204202
- Cited by
- 31
- References
- 5
- FWCI
- 3.09
- Type
- conference paper
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.