Improved RawNet with Feature Map Scaling for Text-Independent Speaker Verification Using Raw Waveforms
Open accessThe DOI matches its Crossref record; 34 of the 34 references it cites were found in a registry or an index.
Abstract
Recent advances in deep learning have facilitated the design of speaker verification systems that directly input raw waveforms.For example, RawNet [1] extracts speaker embeddings from raw waveforms, which simplifies the process pipeline and demonstrates competitive performance.In this study, we improve RawNet by scaling feature maps using various methods.The proposed mechanism utilizes a scale vector that adopts a sigmoid non-linear function.It refers to a vector with dimensionality equal to the number of filters in a given feature map.Using a scale vector, we propose to scale the feature map multiplicatively, additively, or both.In addition, we investigate replacing the first convolution layer with the sinc-convolution layer of SincNet.Experiments performed on the VoxCeleb1 evaluation dataset demonstrate the effectiveness of the proposed methods, and the best performing system reduces the equal error rate by half compared to the original RawNet.Expanded evaluation results obtained using the VoxCeleb1-E and VoxCeleb-H protocols marginally outperform existing state-ofthe-art systems.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Originating from [39] and based on RawNet [40] and graph attention networks, AASIST [41] is representative of progress in the use of E2E architectures.”
From ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wild (2023) · cites this paper as [40]
In the citing paper
“Jung et al. further introduced RawNet2 [121], which replaces CNN with SincNet to better capture aggregated frequency responses. Experimental results demonstrate that RawNet2 outperforms the original RawNet across multiple benchmark tests”
From Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead (2025) · cites this paper as [121]
In the citing paper
“The purpose of FMS is to modify the size of a given feature map independently, the output of the residual block, to obtain a more discriminative representation and improve the performance and generalization of the model, with the advantages of reducing model parameters and computation.”
From Voice Deepfake Detection Using the Self-Supervised Pre-Training Model HuBERT (2023) · cites this paper as [21]
In the citing paper
“However, the limitations are that FMS uses the same scaling vector for addition and multiplication, can only add values between 0 and 1 during addition, and has difficulty optimizing addition and multiplication operations simultaneously when performing multiplication.”
From Voice Deepfake Detection Using the Self-Supervised Pre-Training Model HuBERT (2023) · cites this paper as [21]
Citations
Cites
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404
Exploiting Generative AI to Scale up Intelligent Tutoring Systems
Jakubův, J., Chvalovský, K., Goertzel, Z., Kaliszyk, C., Olšák, M., Piotrowski, B., Schulz, S., Suda, M., & Urban, J. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.itp.2023.19 · Cited by 76,658
PyTorch: An Imperative Style, High-Performance Deep Learning Library
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al. (2019) · arXiv · doi:10.48550/arxiv.1912.01703 · Cited by 31,473
CBAM: Convolutional Block Attention Module
Woo, S., Park, J., Lee, J.-Y., & Kweon, I. S. (2018) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-01234-2_1 · Cited by 25,853
Neural Machine Translation by Jointly Learning to Align and Translate
Bahdanau, D., Cho, K., & Bengio, Y. (2014) · arXiv · doi:10.48550/arxiv.1409.0473 · Cited by 16,213
Squeeze-and-Excitation Networks
Hu, J., Shen, L., Albanie, S., Sun, G., & Wu, E. (2020) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2913372 · Cited by 12,157
Identity mappings in deep residual networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_38 · Cited by 10,741
Identity Mappings in Deep Residual Networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · arXiv · doi:10.48550/arxiv.1603.05027 · Cited by 10,665
A Discriminative Feature Learning Approach for Deep Face Recognition
Wen, Y., Zhang, K., Li, Z., & Qiao, Y. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46478-7_31 · Cited by 4,077
X-Vectors: Robust DNN Embeddings for Speaker Recognition
Snyder, D., Garcia-Romero, D., Sell, G., Povey, D., & Khudanpur, S. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8461375 · Cited by 2,620
Cited by
Items 101–113 of 113
A Synthetic Speech Detection Model Combining Local-Global Dependency
Song, J., Zhang, Y., & Yuan, W. (2025) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2025.069918 · Cited by 1
Tri-View Collaborative Graph Learning for Robust Deepfake Speech Detection
Zhang, K., Zhao, Y., Hua, Z., Zhang, Y., Liao, Q., & Jiang, W. Y. (2026) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2026.3660668 · Cited by 1
A Robust Region-Aware Framework for Audio Forgery Localization
Luo, J., Wang, H., Liu, H., Hu, S. C., & Wang, K. (2026) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2026.3661237
Retrieval-Based Deep Difference Modeling for Audio Deepfake Detection
Liu, R., Zhang, J., & Li, H. (2026) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2026.3677607
Human assisted correction for speaker diarization of an incremental collection of documents
Prokopalo, Y. (2022) · theses.fr (ABES) · doi:10.70675/3d5bd7c8zb851z4f15zb277z08020c5b1683
End-to-End Modeling for Speech Spoofing and Deepfake Detection
Tak, H. (2023) · theses.fr (ABES) · doi:10.70675/aa9fb6cbz3f6az45bcz933fzf95d89893e59 · Cited by 4
Hanui: Harnessing Distributional Discrepancies for Singing Voice Deepfake Detection
Um, S., Kim, D., & Kang, H. C. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11461904
Patil, P., Jeevannavar, A., Chikkmath, S., & Nirmala, S. R. (2026) · IEEE International Conference for Convergence in Computing Technology (I3CTCON) · doi:10.1109/i3ctcon68242.2026.11508028
Enhancing Speaker Representations via Graph Attention Networks and Contrastive Learning
Bidargaddi, A. P., D, D. A., & Prasanna, S. (2025) · IEEE Conference on Engineering Informatics (ICEI) · doi:10.1109/icei68356.2025.11603690
Long-Term Mask and Short-Term Correlation Learning Framework for Speaker Verification
Li, W., Chen, C., Yang, H., & Wang, L. (2026) · Circuits Systems and Signal Processing · doi:10.1007/s00034-026-03712-1
Scenefake Audio Detection Using Amalgmation of Hybrid Linear Cepstral Features and CNN
Sharma, R., & Dua, M. (2026) · International Conference on Power, Control and Computing Technologies (ICPC2T) · doi:10.1109/icpc2t68221.2026.11646425
Nguyen, T. A., Dinh-Xuan, H., Doan, T.-P., & Jung, S. (2026) · IEEE Access · doi:10.1109/access.2026.3732510
Acoustic Microcues: Breath and Phoneme-Level Signals for Detecting Audio Deepfakes
Han, W., Palasamudram, A., Fullwood, J., & Monrose, F. (2026) · doi:10.1145/3840473.3840514
Metadata
- Authors
- Jung, J.-W., Kim, S.-B., Shim, H.-J., Kim, J.-H., & Yu, H.-J.
- Affiliations
- University of Seoul (KR)
- Corresponding author
- Yu, Ha-Jin
- Venue
- Interspeech 2020
- Publisher
- ISCA
- Published
- 2020-10-25
- Pages
- 1496-1500
- ISSN
- 2308-457X (ISSN-L)
- OpenAlex
- W3096084197
- Cited by
- 109
- References
- 30
- Citations by year
- 2026: 10 · 2025: 16 · 2024: 26 · 2023: 19 · 2022: 21 · 2021: 15 · 2020: 2
- FWCI
- 5.44
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Interspeech (USB) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- text-independent speaker verification, SincNet, VoxCeleb1, speaker embedding
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.