The CCF AATC 2025: Speech Restoration Challenge

2025-09-16

AI Summary

Real-world speech communication is often degraded by multiple concurrent distortions—including non-stationary noise, reverberation, codec compression (e.g., MP3), and secondary artifacts introduced by upstream enhancement models—yet existing methods suffer from limited performance due to isolated modeling of individual degradations. This paper proposes the first end-to-end framework jointly modeling three canonical degradation types: non-stationary noise suppression, dereverberation, decompression, and artifact mitigation, implemented via a unified multi-task learning architecture. We construct a large-scale speech dataset featuring realistic composite degradations and design a comprehensive evaluation protocol balancing perceptual quality, objective metrics (PESQ/STOI), and computational efficiency. Additionally, we incorporate model complexity analysis to advance practical, system-level deployment of speech enhancement. Experiments demonstrate that our method significantly outperforms single-distortion baseline models in both speech quality and intelligibility.

Technology Category

• Natural Language Processing: Speech
• Machine Learning: Multimodal Learning
• Search and Optimization: Non-convex Optimization

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Real-world speech communication is often hampered by a variety of distortions that degrade quality and intelligibility. While many speech enhancement algorithms target specific degradations like noise or reverberation, they often fall short in realistic scenarios where multiple distortions co-exist and interact. To spur research in this area, we introduce the Speech Restoration Challenge as part of the China Computer Federation (CCF) Advanced Audio Technology Competition (AATC) 2025. This challenge focuses on restoring speech signals affected by a composite of three degradation types: (1) complex acoustic degradations including non-stationary noise and reverberation; (2) signal-chain artifacts such as those from MP3 compression; and (3) secondary artifacts introduced by other pre-processing enhancement models. We describe the challenge's background, the design of the task, the comprehensive dataset creation methodology, and the detailed evaluation protocol, which assesses both objective performance and model complexity. Homepage: https://ccf-aatc.org.cn/.

Problem

• Addressing multiple coexisting speech distortions
• Restoring speech affected by composite degradations
• Evaluating enhancement models on realistic scenarios

Innovation

• Composite degradation restoration challenge
• Multiple distortion types integration
• Objective and complexity evaluation protocol