# Hyperdimensional Intelligent Sensing for Efficient Real-Time Audio Processing on Extreme Edge

arXiv: https://arxiv.org/abs/2502.10718 · HTML: https://mlsanity.com/paper/2502.10718/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 26 references that are in the map.
- First posted on arXiv 15 Feb 2025.

## Builds on (most cited first, 7 of 7)

- [Wav2CLIP: Learning Robust Audio Representations From CLIP](https://mlsanity.com/paper/2110.11499/) (2021)
- [HTS-AT: A Hierarchical Token-Semantic Audio Transformer for Sound Classification and Detection](https://mlsanity.com/paper/2202.00874/) (2022)
- [Constrained Few-shot Class-incremental Learning](https://mlsanity.com/paper/2203.16588/) (2022)
- [MAE-AST: Masked Autoencoding Audio Spectrogram Transformer](https://mlsanity.com/paper/2203.16691/) (2022)
- [GraphHD: Efficient graph classification using hyperdimensional computing](https://mlsanity.com/paper/2205.07826/) (2022)
- [PISA: A Binary-Weight Processing-In-Sensor Accelerator for Edge Image Processing](https://mlsanity.com/paper/2202.09035/) (2022)
- [Precision and accuracy of acoustic gunshot location in an urban environment](https://mlsanity.com/paper/2108.07377/) (2021)

## Built on by (most cited first, 3 of 3)

- [DecoHD: Decomposed Hyperdimensional Classification under Extreme Memory Budgets](https://mlsanity.com/paper/2511.03911/) (2025)
- [LogHD: Robust Compression of Hyperdimensional Classifiers via Logarithmic Class-Axis Reduction](https://mlsanity.com/paper/2511.03938/) (2025)
- [FusionSense: Tri-Stage Near-Sensor Learning for Runtime-Adaptive Multimodal Edge Intelligence](https://mlsanity.com/paper/2605.22868/) (2026)

## Abstract (arXiv)

The escalating challenges of managing vast sensor-generated data, particularly in audio applications, necessitate innovative solutions. Current systems face significant computational and storage demands, especially in real-time applications like gunshot detection systems (GSDS), and the proliferation of edge sensors exacerbates these issues. This paper proposes a groundbreaking approach with a near-sensor model tailored for intelligent audio-sensing frameworks. Utilizing a Fast Fourier Transform (FFT) module, convolutional neural network (CNN) layers, and HyperDimensional Computing (HDC), our model excels in low-energy, rapid inference, and online learning. It is highly adaptable for efficient ASIC design implementation, offering superior energy efficiency compared to conventional embedded CPUs or GPUs, and is compatible with the trend of shrinking microphone sensor sizes. Comprehensive evaluations at both software and hardware levels underscore the model's efficacy. Software assessments through detailed ROC curve analysis revealed a delicate balance between energy conservation and quality loss, achieving up to 82.1% energy savings with only 1.39% quality loss. Hardware evaluations highlight the model's commendable energy efficiency when implemented via ASIC design, especially with the Google Edge TPU, showcasing its superiority over prevalent embedded CPUs and GPUs.