一个模型搞定所有延迟要求，语音增强不再需要分别训练
不同实时语音应用对延迟要求天差地别——视频通话要低延迟，语音助手可以稍高——以前得为每种场景单独训练一个模型。这篇论文用一个模型搞定所有：算法延迟靠可配置的“前瞻帧”调节，计算延迟靠“早退机制”控制，训练时用两阶段策略弥补灵活性和性能的差距。简单说，就是同一个模型能根据场景自动调整响应速度，不用再重复训练。它不是你明天能用上的，但指明了语音增强模型走向通用化的方向。
📄 原文摘要(英文)
Different real-time speech applications impose distinct latency budgets, often requiring separately trained enhancement models for each scenario. In this paper, we propose a one-for-all, real-time universal speech enhancement model that provides explicit control over both algorithmic and computational latency. Algorithmic latency is flexibly adjusted via configurable look-ahead frames. To avoid learning inefficiency caused by varying padding configurations, we introduce parallel convolutional layers corresponding to different look-ahead settings. Computational latency is controlled through an early-exit mechanism, enabling inference at different network depths. To narrow the performance gap between specialized and flexible models, we propose a two-stage training strategy with a shared-to-multiple decoder transition. Overall, the proposed framework enables a single model to be deployed across diverse latency budgets without retraining separate models.