arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2011.06195cs.CLcs.AIcs.LG
收藏

半监督式语音语义理解

Towards Semi-Supervised Semantics Understanding from Speech

MIT CSAIL(麻省理工学院计算机科学与人工智能实验室);Amazon AI(亚马逊人工智能)

Cheng-I Lai, Jin Cao, Sravan Bodapati, Shang-Wen Li

摘要:

最近关于口语语言理解(SLU)的许多工作至少在以下三种方式中的一种方面存在不足:模型在Oracle文本输入上进行训练,忽略了自动语音识别(ASR)输出,模型被训练为仅预测意图而没有槽值,或者模型在大量内部数据上进行训练。我们提出了一个干净和通用的框架,直接从语音中学习语义,并从转录的语音中进行半监督来解决这些问题。我们的框架建立在预训练的端到端(E2E)ASR和自监督语言模型(如BERT)基础上,并在有限的目标SLU语料库上进行微调。同时,我们确定了两个不充分的设置下,SLU模型进行了测试:噪声鲁棒性和E2E语义评估。我们测试了所提出的框架下现实的环境噪声和一个新的度量,插槽编辑F1分数,两个公共SLU语料库。实验结果表明,当环境噪声存在且标注语义数据量有限时,以语音为输入的SLU框架在语义理解方面的表现与以甲骨文文本为输入的SLU框架相当。

英文摘要:

Much recent work on Spoken Language Understanding (SLU) falls short in at least one of three ways: models were trained on oracle text input and neglected the Automatics Speech Recognition (ASR) outputs, models were trained to predict only intents without the slot values, or models were trained on a large amount of in-house data. We proposed a clean and general framework to learn semantics directly from speech with semi-supervision from transcribed speech to address these. Our framework is built upon pretrained end-to-end (E2E) ASR and self-supervised language models, such as BERT, and fine-tuned on a limited amount of target SLU corpus. In parallel, we identified two inadequate settings under which SLU models have been tested: noise-robustness and E2E semantics evaluation. We tested the proposed framework under realistic environmental noises and with a new metric, the slots edit F1 score, on two public SLU corpora. Experiments show that our SLU framework with speech as input can perform on par with those with oracle text as input in semantics understanding, while environmental noises are present, and a limited amount of labeled semantics data is available.

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。

↑