# Recurrent Neural Networks for Polyphonic Sound Event Detection in Real Life Recordings

arXiv: https://arxiv.org/abs/1604.00861 · HTML: https://mlsanity.com/paper/1604.00861/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Apr 2016.

## Built on by (most cited first, 8 of 8)

- [Dense-Localizing Audio-Visual Events in Untrimmed Videos: A Large-Scale Benchmark and Baseline](https://mlsanity.com/paper/2303.12930/) (2023)
- [The iNaturalist Sounds Dataset](https://mlsanity.com/paper/2506.00343/) (2025)
- [Benchmarking of eight recurrent neural network variants for breath phase and adventitious sound detection on a self-developed open-access lung sound database-HF_Lung_V1](https://mlsanity.com/paper/2102.03049/) (2021)
- [Decompose the Sounds and Pixels, Recompose the Events](https://mlsanity.com/paper/2112.11547/) (2021)
- [Differentiable short-time Fourier transform with respect to the hop length](https://mlsanity.com/paper/2308.02421/) (2023)
- [Environment Transfer for Distributed Systems](https://mlsanity.com/paper/2101.01863/) (2021)
- [Transferring Voice Knowledge for Acoustic Event Detection: An Empirical Study](https://mlsanity.com/paper/2110.03174/) (2021)
- [Metamorphic Testing for Audio Content Moderation Software](https://mlsanity.com/paper/2509.24215/) (2025)

## Abstract (arXiv)

In this paper we present an approach to polyphonic sound event detection in real life recordings based on bi-directional long short term memory (BLSTM) recurrent neural networks (RNNs). A single multilabel BLSTM RNN is trained to map acoustic features of a mixture signal consisting of sounds from multiple classes, to binary activity indicators of each event class. Our method is tested on a large database of real-life recordings, with 61 classes (e.g. music, car, speech) from 10 different everyday contexts. The proposed method outperforms previous approaches by a large margin, and the results are further improved using data augmentation techniques. Overall, our system reports an average F1-score of 65.5% on 1 second blocks and 64.7% on single frames, a relative improvement over previous state-of-the-art approach of 6.8% and 15.1% respectively.