Lightweight DNN for Full-Band Speech Denoising on Mobile Devices: Exploiting Long and Short Temporal Patterns
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916935802814464 |
|---|---|
| author | Drossos, Konstantinos Heikkinen, Mikko Tsiaflakis, Paschalis |
| author_facet | Drossos, Konstantinos Heikkinen, Mikko Tsiaflakis, Paschalis |
| contents | Speech denoising (SD) is an important task of many, if not all, modern signal processing chains used in devices and for everyday-life applications. While there are many published and powerful deep neural network (DNN)-based methods for SD, few are optimized for resource-constrained platforms such as mobile devices. Additionally, most DNN-based methods for SD are not focusing on full-band (FB) signals, i.e. having 48 kHz sampling rate, and/or low latency cases. In this paper we present a causal, low latency, and lightweight DNN-based method for full-band SD, leveraging both short and long temporal patterns. The method is based on a modified UNet architecture employing look-back frames, temporal spanning of convolutional kernels, and recurrent neural networks for exploiting short and long temporal patterns in the signal and estimated denoising mask. The DNN operates on a causal frame-by-frame basis taking as an input the STFT magnitude, utilizes inverted bottlenecks inspired by MobileNet, employs causal instance normalization for channel-wise normalization, and achieves a real-time factor below 0.02 when deployed on a modern mobile phone. The proposed method is evaluated using established speech denoising metrics and publicly available datasets, demonstrating its effectiveness in achieving an (SI-)SDR value that outperforms existing FB and low latency SD methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_05079 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Lightweight DNN for Full-Band Speech Denoising on Mobile Devices: Exploiting Long and Short Temporal Patterns Drossos, Konstantinos Heikkinen, Mikko Tsiaflakis, Paschalis Audio and Speech Processing Machine Learning Sound Signal Processing Speech denoising (SD) is an important task of many, if not all, modern signal processing chains used in devices and for everyday-life applications. While there are many published and powerful deep neural network (DNN)-based methods for SD, few are optimized for resource-constrained platforms such as mobile devices. Additionally, most DNN-based methods for SD are not focusing on full-band (FB) signals, i.e. having 48 kHz sampling rate, and/or low latency cases. In this paper we present a causal, low latency, and lightweight DNN-based method for full-band SD, leveraging both short and long temporal patterns. The method is based on a modified UNet architecture employing look-back frames, temporal spanning of convolutional kernels, and recurrent neural networks for exploiting short and long temporal patterns in the signal and estimated denoising mask. The DNN operates on a causal frame-by-frame basis taking as an input the STFT magnitude, utilizes inverted bottlenecks inspired by MobileNet, employs causal instance normalization for channel-wise normalization, and achieves a real-time factor below 0.02 when deployed on a modern mobile phone. The proposed method is evaluated using established speech denoising metrics and publicly available datasets, demonstrating its effectiveness in achieving an (SI-)SDR value that outperforms existing FB and low latency SD methods. |
| title | Lightweight DNN for Full-Band Speech Denoising on Mobile Devices: Exploiting Long and Short Temporal Patterns |
| topic | Audio and Speech Processing Machine Learning Sound Signal Processing |
| url | https://arxiv.org/abs/2509.05079 |