基于MLM数据增强的ASR与NLU流水线系统:面向STOP低资源挑战
The Pipeline System of ASR and NLU with MLM-based Data Augmentation toward STOP Low-resource Challenge
AI总结:
该系统采用ASR与NLU流水线,通过微调Whisper和BART,结合MLM数据增强与检索增强,在STOP低资源挑战中取得提醒/天气领域平均EM 69.15,获第一名。
摘要:
本文描述了我们在口语理解大挑战赛(ICASSP信号处理大挑战赛2023的一部分)中低资源领域自适应赛道(赛道3)的系统。在该赛道中,我们采用了ASR和NLU的流水线方法。对于ASR,我们针对每个领域通过上采样微调Whisper。对于NLU,我们在所有赛道3数据上微调BART,然后在低资源领域数据上微调。我们应用了基于掩码语言模型(MLM)的数据增强,其中部分输入令牌和相应的目标标签使用MLM替换。我们还应用了基于检索的方法,模型输入通过相似训练样本进行增强。最终,我们在提醒/天气领域取得了精确匹配(EM)准确率63.3/75.0(平均:69.15),并在挑战赛中获得了第一名。
英文摘要:
This paper describes our system for the low-resource domain adaptation track (Track 3) in Spoken Language Understanding Grand Challenge, which is a part of ICASSP Signal Processing Grand Challenge 2023. In the track, we adopt a pipeline approach of ASR and NLU. For ASR, we fine-tune Whisper for each domain with upsampling. For NLU, we fine-tune BART on all the Track3 data and then on low-resource domain data. We apply masked LM (MLM) -based data augmentation, where some of input tokens and corresponding target labels are replaced using MLM. We also apply a retrieval-based approach, where model input is augmented with similar training samples. As a result, we achieved exact match (EM) accuracy 63.3/75.0 (average: 69.15) for reminder/weather domain, and won the 1st place at the challenge.
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。