mlsanityData to 14 Oct 2026

TinyMyo: a Tiny Foundation Model for Flexible EMG Signal Processing at the Edge

arXiv 2512.15729 ↗First posted 5 Dec 2025eess.SP cs.AI cs.HCcited by 7

Not on the map. Why?

Abstract (arXiv)

Objective: Surface electromyography (EMG) is a non-invasive sensing modality widely used in biomechanics, rehabilitation, prosthetic control, and human-machine interfaces. Despite decades of use, achieving robust generalization across subjects, recording systems, and acquisition protocols remains challenging. While foundation models (FMs) are gaining traction for EMG, existing approaches remain limited to single downstream tasks and lack deployability on embedded platforms. This work addresses these limitations. Methods: We present TinyMyo, a lightweight FM based on a Transformer encoder architecture. The model is pre-trained in a self-supervised manner using masked reconstruction on publicly available datasets. With only 3.6M parameters, TinyMyo is designed to support multiple downstream tasks through minimal task-specific head adaptations. Results: We demonstrate generalization across hand gesture classification, hand kinematic regression, speech production and speech recognition, with performance comparable to or surpassing the state of the art (SoA), and model size below 5M parameters. We achieve SoA results compared to previous FM-based works on the NinaPro DB5 (89.4%), UCI-EMG (97.56%), and EPN-612 (96.74%) datasets. We demonstrate the first-time deployment of an EMG FM on an ultra-low power microcontroller (GAP9), with an inference time of 0.785 s, energy of 44.91 mJ and power envelope of 57.18 mW. Conclusion: TinyMyo demonstrates that compact, self-supervised EMG FM can guarantee strong generalization across multiple downstream tasks while remaining compatible with low-power edge devices. Significance: TinyMyo is the first EMG FM for ultra-low power edge devices, enabling scalable and energy-efficient sensing for motor intent decoding, neuromuscular assessment, and biosignal driven human-machine interaction.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 31 of its 67 references that are in the map. First posted on arXiv 5 Dec 2025.

Builds on

31 of 67 references.

References · 31

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Scaling Laws for Neural Language Modelscited by 6,586
  3. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,718
  4. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  5. 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,905
  6. 2022A Time Series is Worth 64 Words: Long-term Forecasting with Transformerscited by 2,039
  7. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
  8. 2021SimMIM: A Simple Framework for Masked Image Modelingcited by 1,153
  9. 2020On Layer Normalization in the Transformer Architecturecited by 922
  10. 2024MOMENT: A Family of Open Time-series Foundation Modelscited by 482
  11. 2021SpeechBrain: A General-Purpose Speech Toolkitcited by 406
  12. 2024Large Brain Model for Learning Generic Representations with Tremendous EEG Data in BCIcited by 327
19 more
  1. 2021I-BERT: Integer-only BERT Quantizationcited by 255
  2. 2023EfficientViT: Memory Efficient Vision Transformer with Cascaded Group Attentioncited by 172
  3. 2022A Survey on Masked Autoencoder for Self-supervised Learning in Vision and Beyondcited by 58
  4. 2020Inter-database validation of a deep learning approach for automatic sleep scoringcited by 38
  5. 2023Channel Vision Transformers: An Image Is Worth 1 x 16 x 16 Wordscited by 34
  6. 2025LUNA: Efficient and Topology-Agnostic Foundation Model for EEG Signal Analysiscited by 26
  7. 2020Digital Voicing of Silent Speechcited by 24
  8. 2024emg2pose: A Large and Diverse Benchmark for Surface Electromyographic Hand Pose Estimationcited by 22
  9. 2025FEMBA: Efficient and Scalable EEG Analysis with a Bidirectional Mamba Foundation Modelcited by 16
  10. 2021An Improved Model for Voicing Silent Speechcited by 15
  11. 2025CEReBrO: Compact Encoder for Representations of Brain Oscillations Using Efficient Alternating Attentioncited by 14
  12. 2025PhysioWave: A Multi-Scale Wavelet-Transformer for Physiological Signal Representationcited by 12
  13. 2025Towards Robust Multimodal Physiological Foundation Models: Handling Arbitrary Missing Modalitiescited by 11
  14. 2024A Cross-Modal Approach to Silent Speech with LLM-Enhanced Recognitioncited by 9
  15. 2024OTIS: Learning High-Quality Time Series Features With Tiny Encoderscited by 8
  16. 2023Siracusa: A 16 nm Heterogenous RISC-V SoC for Extended Reality with At-MRAM Neural Enginecited by 5
  17. 2024Intelligent Learning Rate Distribution to reduce Catastrophic Forgetting in Transformerscited by 4
  18. 2024Diff-ETS: Learning a Diffusion Probabilistic Model for Electromyography-to-Speech Conversioncited by 4
  19. 2025WaveFormer: A Lightweight Transformer Model for sEMG-based Gesture Recognitioncited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 7.

7 papers.

  1. 2026EgoEMG: A Multimodal Egocentric Dataset with Bilateral EMG and Vision for Hand Pose Estimationcited by 3
  2. 2026A Montage-Agnostic Encoder for Calibration-Light Cross-User Gesture Recognition from Surface Electromyographycited by 2
  3. 2026PanLUNA: An Efficient and Robust Query-Unified Multimodal Model for Edge Biosignal Intelligencecited by 0
  4. 2026EMGBlend: Heterogeneity-Aware Self-Supervised Pretraining for Gesture and Force Decodingcited by 0
  5. 2026MyoCodec: A Streaming Neural Codec for Electromyographycited by 0
  6. 2026EMG-FM-Bench: A Comprehensive Benchmark for Foundation Model Transfer and Adaptation on Electromyographycited by 0
  7. 2026EVFormer: An Egocentric Vision-EMG Bidirectional Attention Model for Bimanual Hand Pose Estimationcited by 0