arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.09681eess.AScs.SD
收藏

通过平均领域专家实现端到端ASR的持续学习

Continual Learning for End-to-End ASR by Averaging Domain Experts

JP Morgan Chase & Co.(摩根大通公司)

Peter Plantinga, Jaekwon Yoo, Chandra Dhir

AI总结:

提出一种通过线性插值多个领域专家模型参数的持续学习策略,在端到端ASR中仅需单一模型即可避免灾难性遗忘,并在所有测试数据上保持良好性能。

摘要:

端到端自动语音识别的持续学习必须应对诸多困难。微调策略往往会损失在已见数据上的性能,这一过程被称为灾难性遗忘。另一方面,冻结参数并附加可调参数的策略必须维护多个模型。我们提出了一种仅维护单个模型进行推理并避免灾难性遗忘的策略。我们的实验表明,对多个模型的参数进行简单的线性插值,每个模型都从同一个通用模型微调而来,可以生成一个在所有测试数据上表现良好的单一模型。在我们的实验中,我们选择了两个在大型数据集上预训练的开源端到端语音识别模型,并在3个独立的数据集上对它们进行微调:SGPISpeech、CORAAL和DiPCo。所提出的领域专家平均模型在所有测试数据上表现良好,并且在原始训练领域的数据上几乎没有性能损失。

英文摘要:

Continual learning for end-to-end automatic speech recognition has to contend with a number of difficulties. Fine-tuning strategies tend to lose performance on data already seen, a process known as catastrophic forgetting. On the other hand, strategies that freeze parameters and append tunable parameters must maintain multiple models. We suggest a strategy that maintains only a single model for inference and avoids catastrophic forgetting. Our experiments show that a simple linear interpolation of several models' parameters, each fine-tuned from the same generalist model, results in a single model that performs well on all tested data. For our experiments we selected two open-source end-to-end speech recognition models pre-trained on large datasets and fine-tuned them on 3 separate datasets: SGPISpeech, CORAAL, and DiPCo. The proposed average of domain experts model performs well on all tested data, and has almost no loss in performance on data from the domain of original training.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑