Federated learning (FL) enables privacy-preserving collaborative training across distributed edge devices, but real deployments involve heterogeneous clients with different processing power, memory capacity, and communication latency, which often increase round duration and system cost. This paper proposes a hardware-aware federated learning framework for emotion recognition on session-partitioned IEMOCAP that integrates hardware profiling, top-K client selection, and adaptive local epochs within a unified training loop. We compare the method against FedAvg, FedProx, and random top-K selection under a non-IID setup and show that, across 50 federated rounds and 5 independent trials, the proposed approach achieves competitive validation accuracy (0.352), reduces total training time by about 36.5% compared to FedAvg, and lowers cumulative communication cost by 40%.

翻译：联邦学习（FL）能够在分布式边缘设备间实现隐私保护的协同训练，但实际部署中涉及异构客户端，其处理能力、内存容量和通信延迟存在差异，这往往增加了训练轮次时长和系统成本。本文提出一种面向硬件的联邦学习框架，用于基于会话分段的IEMOCAP数据集上的情感识别任务。该框架将硬件性能剖析、top-K客户端选择与自适应本地轮次整合至统一的训练循环中。在非独立同分布（non-IID）设置下，我们将该方法与FedAvg、FedProx及随机top-K选择进行对比。实验表明，在50个联邦训练轮次和5次独立试验中，该方法的验证准确率（0.352）具有竞争力，总训练时间较FedAvg减少约36.5%，累积通信成本降低40%。