PolyChirp: Multi-Species Birdsong Classification Using TinyML on Low-Power Acoustic Sensors
Not on the map. Why?
Abstract (arXiv)
Recent progress in the field of TinyML has demonstrated that low-power hardware based on microcontrollers can achieve bird species monitoring in real time based on acoustic sensor data for an entire breeding period on a single battery charge. However, the state of the art on low-power microcontrollers was so far limited to binary classification of a single species. In contrast, real fauna monitoring deployments often target multiple species simultaneously. To address this challenge we develop PolyChirp, an approach combining biological domain expertise, automated dataset curation, neural architecture optimization and novel hardware to achieve multiclass bird species detection in the wild. PolyChirp is based on newly designed tiny multiclass models that leverage recent microcontrollers and hardware acceleration with a neural processing unit (NPU). We evaluate the predictive performance of these models, and we measure their computational performance -- flash footprint, latency, energy consumption -- on common microcontroller hardware. Our results demonstrate that PolyChirp matches or exceeds the TinyChirp architectures retrained under our protocol on single-species detection, and further achieves robust classification of up to 10 species simultaneously (macro F2 up to 0.97), while still fitting the flash, latency and energy budget of a low-power microcontroller sensor. A data-driven front-end redesign additionally makes on-device mel feature extraction 7x to 11x cheaper.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 23 of its 44 references that are in the map. First posted on arXiv 24 Aug 2026.
Builds on
23 of 44 references.
References · 23
- 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,593
- 2021A White Paper on Neural Network Quantizationcited by 472
- 2020MCUNet: Tiny Deep Learning on IoT Devicescited by 273
- 2018Speaker Recognition from Raw Waveform with SincNetcited by 166
- 2020TensorFlow Lite Micro: Embedded Machine Learning on TinyML Systemscited by 165
- 2021MLPerf Tiny Benchmarkcited by 124
- 2018CMSIS-NN: Efficient Neural Network Kernels for Arm Cortex-M CPUscited by 121
- 2021LEAF: A Learnable Frontend for Audio Classificationcited by 69
- 2023Global birdsong embeddings enable superior transfer learning for bioacoustic classificationcited by 51
- 2018Automatic acoustic detection of birds through deep learning: the first Bird Audio Detection challengecited by 44
- 2016Trainable Frontend For Robust and Far-Field Keyword Spottingcited by 33
- 2019End-to-End Environmental Sound Classification using a 1D Convolutional Neural Networkcited by 20
11 more
- 2025Benchmarking Ultra-Low-Power μNPUscited by 6
- 2023U-TOE: Universal TinyML On-board Evaluation Toolkit for Low-Power IoTcited by 5
- 2022Keyword Spotting System and Evaluation of Pruning and Quantization Methods on Low-power Edge Microcontrollerscited by 4
- 2024TinyChirp: Bird Song Recognition Using TinyML Models on Low-power Wireless Acoustic Sensorscited by 4
- 2024MicroFlow: An Efficient Rust-Based Inference Engine for TinyMLcited by 3
- 2025msf-CNN: Patch-based Multi-Stage Fusion with Convolutional Neural Networks for TinyMLcited by 3
- 2024Multispecies bird sound recognition using a fully convolutional neural networkcited by 2
- 2025Enabling Multi-Species Bird Classification on Low-Power Bioacoustic Loggerscited by 2
- 2024TinyBird-ML: An ultra-low Power Smart Sensor Node for Bird Vocalization Analysis and Syllable Classificationcited by 1
- 2025TinyD'ej`aVu: Smaller RAM and Faster Inference with Neural Networks on MCUs for Sensor Data Streamscited by 1
- 2025Ariel-ML: Computing Parallelization with Embedded Rust for Neural Networks on Heterogeneous Multi-core Microcontrollerscited by 1
Built on by
Not cited yet.