# SaD: A Scenario-Aware Discriminator for Speech Enhancement

## TL;DR

This paper introduces a scenario-aware discriminator for speech enhancement that captures scene-specific features and improves quality assessment, leading to better performance across diverse scenarios without changing generator architectures.

## Contribution

The proposed discriminator uniquely incorporates scene-specific features and frequency-domain division, enhancing speech quality evaluation in GAN-based models.

## Key findings

- Effective adaptation to various generator architectures
- Improved speech enhancement performance across scenarios
- Validated on multiple datasets and models

## Abstract

Generative adversarial network-based models have shown remarkable performance in the field of speech enhancement. However, the current optimization strategies for these models predominantly focus on refining the architecture of the generator or enhancing the quality evaluation metrics of the discriminator. This approach often overlooks the rich contextual information inherent in diverse scenarios. In this paper, we propose a scenario-aware discriminator that captures scene-specific features and performs frequency-domain division, thereby enabling a more accurate quality assessment of the enhanced speech generated by the generator. We conducted comprehensive experiments on three representative models using two publicly available datasets. The results demonstrate that our method can effectively adapt to various generator architectures without altering their structure, thereby unlocking further performance gains in speech enhancement across different scenarios.