# PolyChirp: Multi-Species Birdsong Classification Using TinyML on Low-Power Acoustic Sensors

arXiv: https://arxiv.org/abs/2608.23101 · HTML: https://mlsanity.com/paper/2608.23101/

## 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 (most cited first, 23 of 23)

- [Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference](https://mlsanity.com/paper/1712.05877/) (2017)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [MCUNet: Tiny Deep Learning on IoT Devices](https://mlsanity.com/paper/2007.10319/) (2020)
- [Speaker Recognition from Raw Waveform with SincNet](https://mlsanity.com/paper/1808.00158/) (2018)
- [TensorFlow Lite Micro: Embedded Machine Learning on TinyML Systems](https://mlsanity.com/paper/2010.08678/) (2020)
- [MLPerf Tiny Benchmark](https://mlsanity.com/paper/2106.07597/) (2021)
- [CMSIS-NN: Efficient Neural Network Kernels for Arm Cortex-M CPUs](https://mlsanity.com/paper/1801.06601/) (2018)
- [LEAF: A Learnable Frontend for Audio Classification](https://mlsanity.com/paper/2101.08596/) (2021)
- [Global birdsong embeddings enable superior transfer learning for bioacoustic classification](https://mlsanity.com/paper/2307.06292/) (2023)
- [Automatic acoustic detection of birds through deep learning: the first Bird Audio Detection challenge](https://mlsanity.com/paper/1807.05812/) (2018)
- [Trainable Frontend For Robust and Far-Field Keyword Spotting](https://mlsanity.com/paper/1607.05666/) (2016)
- [End-to-End Environmental Sound Classification using a 1D Convolutional Neural Network](https://mlsanity.com/paper/1904.08990/) (2019)
- [Benchmarking Ultra-Low-Power μNPUs](https://mlsanity.com/paper/2503.22567/) (2025)
- [U-TOE: Universal TinyML On-board Evaluation Toolkit for Low-Power IoT](https://mlsanity.com/paper/2306.14574/) (2023)
- [Keyword Spotting System and Evaluation of Pruning and Quantization Methods on Low-power Edge Microcontrollers](https://mlsanity.com/paper/2208.02765/) (2022)
- [TinyChirp: Bird Song Recognition Using TinyML Models on Low-power Wireless Acoustic Sensors](https://mlsanity.com/paper/2407.21453/) (2024)
- [MicroFlow: An Efficient Rust-Based Inference Engine for TinyML](https://mlsanity.com/paper/2409.19432/) (2024)
- [msf-CNN: Patch-based Multi-Stage Fusion with Convolutional Neural Networks for TinyML](https://mlsanity.com/paper/2505.11483/) (2025)
- [Multispecies bird sound recognition using a fully convolutional neural network](https://mlsanity.com/paper/2402.05489/) (2024)
- [Enabling Multi-Species Bird Classification on Low-Power Bioacoustic Loggers](https://mlsanity.com/paper/2509.20103/) (2025)
- [TinyBird-ML: An ultra-low Power Smart Sensor Node for Bird Vocalization Analysis and Syllable Classification](https://mlsanity.com/paper/2407.21486/) (2024)
- [TinyD'ej`aVu: Smaller RAM and Faster Inference with Neural Networks on MCUs for Sensor Data Streams](https://mlsanity.com/paper/2512.09786/) (2025)
- [Ariel-ML: Computing Parallelization with Embedded Rust for Neural Networks on Heterogeneous Multi-core Microcontrollers](https://mlsanity.com/paper/2512.09800/) (2025)

## 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.