标题： FasTUSS：更快的任务感知统一源分离 显示英文标题

标题： FasTUSS: Faster Task-Aware Unified Source Separation

摘要： 时间-频率（TF）双路径模型目前是表现最好的音频源分离网络架构，在语音增强、音乐源分离和电影音频源分离中达到了最先进性能。 尽管它们的参数数量相对较低，但仍需要大量的运算，这意味着执行时间更长。 这个问题由于更大模型的趋势而加剧，这些模型在大量数据上训练以解决更通用的任务，例如最近引入的任务感知统一源分离（TUSS）模型。 TUSS旨在使用一个条件模型解决音频源分离任务，它建立在TF-Locoformer之上，TF-Locoformer是一种结合卷积和注意力层的TF双路径模型。 任务定义以一系列提示的形式出现，这些提示指定了要提取的源的数量和类型。 在本文中，我们分析了TUSS的设计选择，以优化其性能-复杂度权衡。 我们推导出两个更高效的模型，FasTUSS-8.3G和FasTUSS-11.7G，分别将原始模型的运算量减少了81%和73%，平均在所有基准测试中性能下降分别为1.2~dB和0.4~dB。 此外，我们研究了提示条件的影响，推导出一个因果TUSS模型。 显示英文摘要

摘要： Time-Frequency (TF) dual-path models are currently among the best performing audio source separation network architectures, achieving state-of-the-art performance in speech enhancement, music source separation, and cinematic audio source separation. While they are characterized by a relatively low parameter count, they still require a considerable number of operations, implying a higher execution time. This problem is exacerbated by the trend towards bigger models trained on large amounts of data to solve more general tasks, such as the recently introduced task-aware unified source separation (TUSS) model. TUSS, which aims to solve audio source separation tasks using a single, conditional model, is built upon TF-Locoformer, a TF dual-path model combining convolution and attention layers. The task definition comes in the form of a sequence of prompts that specify the number and type of sources to be extracted. In this paper, we analyze the design choices of TUSS with the goal of optimizing its performance-complexity trade-off. We derive two more efficient models, FasTUSS-8.3G and FasTUSS-11.7G that reduce the original model's operations by 81\% and 73\% with minor performance drops of 1.2~dB and 0.4~dB averaged over all benchmarks, respectively. Additionally, we investigate the impact of prompt conditioning to derive a causal TUSS model.