%Aigaion2 BibTeX export from Idiap Publications
%Friday 13 November 2026 04:59:26 PM

@ARTICLE{ElHajal_ARXIV_2026,
                      author = {El Hajal, Karl and Magimai-Doss, Mathew},
                    projects = {Idiap, PaSS},
         mainresearchprogram = {Human-AI Teaming},
                       title = {OLIVE: View-Augmented Latent Prediction with Waveform Reconstruction for Speech SSL},
                     journal = {arXiv},
                        year = {2026},
                         doi = {https://doi.org/10.48550/arXiv.2606.30356},
                    abstract = {We propose Online Latent prediction with Invariant Views and rEconstruction
(OLIVE), a self-supervised speech representation learning framework that jointly
optimizes analysis and synthesis objectives. OLIVE couples masked latent predic-
tion and waveform reconstruction through a functional separation within a single
encoder: reconstruction constrains early encoder features to retain signal-level
information, while masked latent prediction shapes later contextual representations for robust downstream performance, with waveform view augmentations
defining the invariances they learn. To the best of our knowledge, OLIVE is the
first speech SSL model that directly reconstructs waveforms, retains its vocoder
after pre-training, and remains competitive on discriminative tasks. Adding reconstruction largely preserves the speech recognition performance of masked latent
prediction alone, while improving waveform reconstruction and speaker tasks.
Adding waveform mixup and gain augmentations provides a controllable way to
shift the representations further toward speaker, generation, and reconstruction
tasks, while remaining competitive on recognition and semantic tasks.}
}