Scholarlink

<p>Automatic speech recognition models require large amounts of speech recordings for training. However, the collection of such data is often cumbersome and leads to privacy concerns. Federated learning has been widely used as an effective decentralized technique that collaboratively learns a shared prediction model while keeping the data local on different clients. Unfortunately, client devices often feature limited computational and communication resources, leading to practical difficulties for large models. In addition, the heterogeneity that characterizes edge devices makes it sub-optimal to generate a single model that fits all of them. Differently from recent literature where multiple models with different architectures are used, we propose using dynamic architectures which, employing early-exit solutions, can adapt their processing (i.e. traversed layers) depending on the input and on the operation conditions. This solution falls in the realm of partial training methods and brings two benefits: <b>❶)</b> a single model is used on a variety of devices and <b>❷)</b> federating the models after local training is straightforward. Experiments on public datasets show that our proposed approach is effective and can be combined with basic federated learning strategies.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Federating dynamic models using early-exit architectures for automatic speech recognition on heterogeneous clients

  • 期刊论文

  • 发表日期：

    2025年10月10日

查看全文
  • Mohamed Nabih Ali,
  • Daniele Falavigna,
  • Alessio Brutti

摘要

Automatic speech recognition models require large amounts of speech recordings for training. However, the collection of such data is often cumbersome and leads to privacy concerns. Federated learning has been widely used as an effective decentralized technique that collaboratively learns a shared prediction model while keeping the data local on different clients. Unfortunately, client devices often feature limited computational and communication resources, leading to practical difficulties for large models. In addition, the heterogeneity that characterizes edge devices makes it sub-optimal to generate a single model that fits all of them. Differently from recent literature where multiple models with different architectures are used, we propose using dynamic architectures which, employing early-exit solutions, can adapt their processing (i.e. traversed layers) depending on the input and on the operation conditions. This solution falls in the realm of partial training methods and brings two benefits: ❶) a single model is used on a variety of devices and ❷) federating the models after local training is straightforward. Experiments on public datasets show that our proposed approach is effective and can be combined with basic federated learning strategies.