# Deep neural network Based Low-latency Speech Separation with Asymmetric analysis-Synthesis Window Pair

arXiv: https://arxiv.org/abs/2106.11794 · HTML: https://mlsanity.com/paper/2106.11794/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Jun 2021.

## Built on by (most cited first, 6 of 6)

- [Iterative autoregression: a novel trick to improve your low-latency speech enhancement model](https://mlsanity.com/paper/2211.01751/) (2022)
- [DPSNN: Spiking Neural Network for Low-Latency Streaming Speech Enhancement](https://mlsanity.com/paper/2408.07388/) (2024)
- [Modulating State Space Model with SlowFast Framework for Compute-Efficient Ultra Low-Latency Speech Enhancement](https://mlsanity.com/paper/2411.02019/) (2024)
- [A Survey of Deep Learning for Complex Speech Spectrograms](https://mlsanity.com/paper/2505.08694/) (2025)
- [Scaling Context Requires Rethinking Attention](https://mlsanity.com/paper/2507.04239/) (2025)
- [Towards Sub-millisecond Latency Real-Time Speech Enhancement Models on Hearables](https://mlsanity.com/paper/2409.18239/) (2024)

## Abstract (arXiv)

Time-frequency masking or spectrum prediction computed via short symmetric windows are commonly used in low-latency deep neural network (DNN) based source separation. In this paper, we propose the usage of an asymmetric analysis-synthesis window pair which allows for training with targets with better frequency resolution, while retaining the low-latency during inference suitable for real-time speech enhancement or assisted hearing applications. In order to assess our approach across various model types and datasets, we evaluate it with both speaker-independent deep clustering (DC) model and a speaker-dependent mask inference (MI) model. We report an improvement in separation performance of up to 1.5 dB in terms of source-to-distortion ratio (SDR) while maintaining an algorithmic latency of 8 ms.