@inproceedings{moradiLightweightOnDeviceAudio2026,

title = {Towards Lightweight On-Device Audio Deepfake Detection Using Squeezeformers},

author = {A. Moradi and Y. Zhu and T. H. Falk},

editor = {Adi K. and Nguena Timo O. and Boulahia-Cuppens N. and Espes D. and Stakhanova N. and Omar M.},

url = {https://www.scopus.com/pages/publications/105046137533?origin=resultslist},

doi = {10.1007/978-3-032-20732-6_24},

isbn = {03029743 (ISSN); 978-303220731-9 (ISBN)},

year  = {2026},

date = {2026-01-01},

booktitle = {Lect. Notes Comput. Sci.},

volume = {16295 LNCS},

pages = {376–389},

publisher = {Springer Science and Business Media Deutschland GmbH},

abstract = {The increasing threat of audio deepfakes necessitates detection mechanisms that can operate in real-time on resource-constrained edge devices. While large-scale systems, such as detectors based on speech foundation models, have demonstrated high accuracy, their computational and memory footprints make them ill-suited for on-device applications. This paper addresses this critical gap by investigating the key factors that influence the performance of lightweight deepfake detection models. We conduct a systematic comparison of model architectures, input feature choices, and data augmentation techniques, evaluating both deepfake detection accuracy and computational complexity across three datasets. Our findings show that with proper modeling choices, a lightweight model can achieve performance comparable to that of a much larger model while being approximately 100× smaller in size. This work provides actionable insights for developing efficient and effective audio deepfake detectors tailored for the constraints of edge computing. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.},

note = {Journal Abbreviation: Lect. Notes Comput. Sci.},

keywords = {Audio DeepFake Detection, Detection mechanism, Edge Computing, Edge detection, Foundation models, High-accuracy, Large scale systems, Large-scale systems, Lightweight, Memory footprint, Performance, Real- time},

pubstate = {published},

tppubtype = {inproceedings}

}

@inproceedings{moradiBenchmarkingFoundationModels2026,

title = {Benchmarking Foundation Models for Cross-Domain Speaker Profiling},

author = {A. Moradi and T. H. Falk},

url = {https://www.scopus.com/pages/publications/105042046048?origin=resultslist},

doi = {10.1109/CAI68641.2026.11536565},

isbn = {979-833156039-3 (ISBN)},

year  = {2026},

date = {2026-01-01},

booktitle = {IEEE Conf. Artif. Intell., CAI},

pages = {78–84},

publisher = {Institute of Electrical and Electronics Engineers Inc.},

abstract = {Speech conveys both linguistic and paralinguistic content. While pre-trained speech foundation models have been widely explored for linguistic tasks, such as speech recognition, and for speaker identification and verification tasks, very limited work has been done to test their usefulness for multi-attribute speaker profiling, i.e., simultaneous prediction of biological sex, age, and height from speech. In this paper, we aim to benchmark the performance of four state-of-the-art self-supervised foundation models, namely, WavLM, Wav2vec2, HuBERT, and XLSR-53, under both within- and cross-domain conditions. Each model is employed as a frozen pre-trained feature extractor, with lightweight task-specific heads trained jointly in a multi-task learning framework, and their feature extraction latency is empirically analyzed to assess practical deployment considerations. Experiments on two datasets show that WavLM achieves consistently strong within- and cross-domain performance for biological sex prediction, while Wav2vec2 and XLSR-53 exhibit more consistent performance for the age and height regression tasks in cross-domain settings. Interpretability analysis based on correlations with key acoustic features show (1) WavLM internal representations correlating highly with pitch and formant frequencies, corroborating the improved performance on biological sex prediction, and (2) XLSR-53 correlating highly with the second formant frequency, corroborating the results obtained for age and height. Overall, our analysis shows that pre-trained speech foundation models could serve as useful tools for cross-domain speaker profiling tasks. While no model stood out as a clear winner across all tested physical traits, future work could explore the use of ensemble methods for improved generalizability. © 2026 IEEE.},

note = {Journal Abbreviation: IEEE Conf. Artif. Intell., CAI},

keywords = {Benchmarking, Continuous speech recognition, Cross-domain, Forecasting, Formant frequency, Foundation models, Learning systems, Linguistics, Multi-attributes, Multi-task learning, Paralinguistic, Performance, Speaker identification, Speaker verification, Speech communication, State of the art, Verification task},

pubstate = {published},

tppubtype = {inproceedings}

}