An overview of neural architectures for self-supervised audio representation learning from masked spectrograms

• URL: http://arxiv.org/abs/2509.18691v1
• Date: Tue, 23 Sep 2025 06:20:41 GMT
• Title: An overview of neural architectures for self-supervised audio representation learning from masked spectrograms
• Authors: Sarthak Yadav, Sergios Theodoridis, Zheng-Hua Tan,
• Abstract summary: Masked spectrogram modeling has emerged as the prominent approach for learning general purpose audio representations.<n>We present a comprehensive overview of the aforementioned research domains, covering masked spectrogram modeling and the neural sequence modeling architectures.
• Score: 22.53614046805595
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: In recent years, self-supervised learning has amassed significant interest for training deep neural representations without labeled data. One such self-supervised learning approach is masked spectrogram modeling, where the objective is to learn semantically rich contextual representations by predicting removed or hidden portions of the input audio spectrogram. With the Transformer neural architecture at its core, masked spectrogram modeling has emerged as the prominent approach for learning general purpose audio representations, a.k.a. audio foundation models. Meanwhile, addressing the issues of the Transformer architecture, in particular the underlying Scaled Dot-product Attention operation, which scales quadratically with input sequence length, has led to renewed interest in recurrent sequence modeling approaches. Among them, Selective structured state space models (such as Mamba) and extended Long Short-Term Memory (xLSTM) are the two most promising approaches which have experienced widespread adoption. While the body of work on these two topics continues to grow, there is currently a lack of an adequate overview encompassing the intersection of these topics. In this paper, we present a comprehensive overview of the aforementioned research domains, covering masked spectrogram modeling and the previously mentioned neural sequence modeling architectures, Mamba and xLSTM. Further, we compare Transformers, Mamba and xLSTM based masked spectrogram models in a unified, reproducible framework on ten diverse downstream audio classification tasks, which will help interested readers to make informed decisions regarding suitability of the evaluated approaches to adjacent applications.

Related papers

• Representation-Regularized Convolutional Audio Transformer for Audio Understanding [53.092757178419355]
  bootstrapping representations from scratch is computationally expensive, often requiring extensive training to converge.<n>We propose the Convolutional Audio Transformer (CAT), a unified framework designed to address these challenges.
  arXiv  Detail & Related papers  (2026-01-29T12:16:19Z)
• Complementary and Contrastive Learning for Audio-Visual Segmentation [74.11434759171199]
  We present Complementary and Contrastive Transformer (CCFormer), a novel framework adept at processing both local and global information.<n>Our method sets new state-of-the-art benchmarks across the S4, MS3 and AVSS datasets.
  arXiv  Detail & Related papers  (2025-10-11T06:36:59Z)
• Scalable Audio-Visual Masked Autoencoders for Efficient Affective Video Facial Analysis [11.373305523732718]
  Affective video facial analysis (AVFA) has emerged as a key research field for building emotion-aware intelligent systems.<n>Masked Autoencoders (MAE) has gained momentum, with growing adaptations in its audio-visual contexts.<n>AVF-MAE++ is a family of audio-visual MAE models designed to efficiently investigate the scaling properties in AVFA.
  arXiv  Detail & Related papers  (2025-09-29T02:53:49Z)
• Harmonizing Visual Representations for Unified Multimodal Understanding and Generation [53.01486796503091]
  We present emphHarmon, a unified autoregressive framework that harmonizes understanding and generation tasks with a shared MAR encoder.<n>Harmon achieves state-of-the-art image generation results on the GenEval, MJHQ30K and WISE benchmarks.
  arXiv  Detail & Related papers  (2025-03-27T20:50:38Z)
• Computation-Efficient Era: A Comprehensive Survey of State Space Models in Medical Image Analysis [8.115549269867403]
  State Space Models (SSMs) have garnered immense interest lately in sequential modeling and visual representation learning. Capitalizing on the advances in computer vision, medical imaging has heralded a new epoch with Mamba models.
  arXiv  Detail & Related papers  (2024-06-05T16:29:03Z)
• TokenUnify: Scaling Up Autoregressive Pretraining for Neuron Segmentation [65.65530016765615]
  We propose a hierarchical predictive coding framework that captures multi-scale dependencies through three complementary learning objectives.<n> TokenUnify integrates random token prediction, next-token prediction, and next-all token prediction to create a comprehensive representational space.<n>We also introduce a large-scale EM dataset with 1.2 billion annotated voxels, offering ideal long-sequence visual data with spatial continuity.
  arXiv  Detail & Related papers  (2024-05-27T05:45:51Z)
• Unveiling the Power of Audio-Visual Early Fusion Transformers with Dense Interactions through Masked Modeling [24.346868432774453]
  Humans possess a remarkable ability to integrate auditory and visual information, enabling a deeper understanding of the surrounding environment. This early fusion of audio and visual cues, demonstrated through cognitive psychology and neuroscience research, offers promising potential for developing multimodal perception models. We address training early fusion architectures by leveraging the masked reconstruction framework, previously successful in unimodal settings, to train audio-visual encoders with early fusion. We propose an attention-based fusion module that captures interactions between local audio and visual representations, enhancing the model's ability to capture fine-grained interactions.
  arXiv  Detail & Related papers  (2023-12-02T03:38:49Z)
• Dynamic Spectrum Mixer for Visual Recognition [17.180863898764194]
  We propose a content-adaptive yet computationally efficient structure, dubbed Dynamic Spectrum Mixer (DSM) DSM represents token interactions in the frequency domain by employing the Cosine Transform. It can learn long-term spatial dependencies with log-linear complexity.
  arXiv  Detail & Related papers  (2023-09-13T04:51:15Z)
• ASiT: Local-Global Audio Spectrogram vIsion Transformer for Event Classification [42.95038619688867]
  ASiT is a novel self-supervised learning framework that captures local and global contextual information by employing group masked model learning and self-distillation. We evaluate our pretrained models on both audio and speech classification tasks, including audio event classification, keyword spotting, and speaker identification.
  arXiv  Detail & Related papers  (2022-11-23T18:21:09Z)
• Masked Frequency Modeling for Self-Supervised Visual Pre-Training [102.89756957704138]
  We present Masked Frequency Modeling (MFM), a unified frequency-domain-based approach for self-supervised pre-training of visual models. MFM first masks out a portion of frequency components of the input image and then predicts the missing frequencies on the frequency spectrum. For the first time, MFM demonstrates that, for both ViT and CNN, a simple non-Siamese framework can learn meaningful representations even using none of the following: (i) extra data, (ii) extra model, (iii) mask token.
  arXiv  Detail & Related papers  (2022-06-15T17:58:30Z)
• Any-to-Many Voice Conversion with Location-Relative Sequence-to-Sequence Modeling [61.351967629600594]
  This paper proposes an any-to-many location-relative, sequence-to-sequence (seq2seq), non-parallel voice conversion approach. In this approach, we combine a bottle-neck feature extractor (BNE) with a seq2seq synthesis module. Objective and subjective evaluations show that the proposed any-to-many approach has superior voice conversion performance in terms of both naturalness and speaker similarity.
  arXiv  Detail & Related papers  (2020-09-06T13:01:06Z)
• Deep Variational Generative Models for Audio-visual Speech Separation [33.227204390773316]
  We propose an unsupervised technique based on audio-visual generative modeling of clean speech. To better utilize the visual information, the posteriors of the latent variables are inferred from mixed speech. Our experiments show that the proposed unsupervised VAE-based method yields better separation performance than NMF-based approaches.
  arXiv  Detail & Related papers  (2020-08-17T10:12:33Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.