arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.10083cs.CLcs.SDeess.AS
收藏

关于语音基础模型在口语理解中的评估

On the Evaluation of Speech Foundation Models for Spoken Language Understanding

Carnegie Mellon University(卡内基梅隆大学);Toyota Technological Institute at Chicago(芝加哥丰田理工学院);ASAPP(ASAPP公司);National Taiwan University(台湾大学)

Siddhant Arora, Ankita Pasad, Chung-Ming Chien, Jionghao Han, Roshan Sharma, Jee-weon Jung, Hira Dhamyal, William Chen, Suwon Shon, Hung-yi Lee, Karen Livescu, Shinji Watanabe

AI总结:

本文通过冻结/微调与不同预测头协议系统评估多种监督式和自监督语音基础模型,发现自监督模型在 SLUE 任务上常具竞争力,并发布 SLUE-PERB 开源工具包与排行榜。

摘要:

Spoken Language Understanding Evaluation(SLUE)基准任务套件最近被提出,以满足在自然语音上对复杂口语理解(SLU)任务(包括分类和序列生成任务)进行开放资源与基准测试的需求。该基准已初步证明,使用预训练语音基础模型(SFM)处理这些 SLU 任务是成功的。然而,研究界仍缺乏对不同 SFM 相对效用的细粒度理解。受此启发,我们提出问题:哪些 SFM 对这些复杂 SLU 任务最有益,以及整合这些 SFM 的最有效方法是什么?为回答这一问题,我们采用多种评估协议对多个监督式和自监督 SFM 进行了广泛评估:(i)冻结 SFM 并搭配轻量级预测头;(ii)冻结 SFM 并搭配复杂预测头;(iii)微调 SFM 并搭配轻量级预测头。尽管监督式 SFM 在多得多的(带标签的)语音识别数据上进行了预训练,但它们并不总是优于自监督 SFM;后者往往表现得至少与监督式 SFM 一样好,有时甚至更好,尤其是在 SLUE 中的序列生成任务上。虽然不存在普遍最优的 SFM 整合方式,但复杂预测头在大多数任务上取得了最佳性能,尽管它会增加推理时间。我们还针对这些任务和建模策略推出了一个开源工具包和性能排行榜 SLUE-PERB。

英文摘要:

The Spoken Language Understanding Evaluation (SLUE) suite of benchmark tasks was recently introduced to address the need for open resources and benchmarking of complex spoken language understanding (SLU) tasks, including both classification and sequence generation tasks, on natural speech. The benchmark has demonstrated preliminary success in using pre-trained speech foundation models (SFM) for these SLU tasks. However, the community still lacks a fine-grained understanding of the comparative utility of different SFMs. Inspired by this, we ask: which SFMs offer the most benefits for these complex SLU tasks, and what is the most effective approach for incorporating these SFMs? To answer this, we perform an extensive evaluation of multiple supervised and self-supervised SFMs using several evaluation protocols: (i) frozen SFMs with a lightweight prediction head, (ii) frozen SFMs with a complex prediction head, and (iii) fine-tuned SFMs with a lightweight prediction head. Although the supervised SFMs are pre-trained on much more speech recognition data (with labels), they do not always outperform self-supervised SFMs; the latter tend to perform at least as well as, and sometimes better than, supervised SFMs, especially on the sequence generation tasks in SLUE. While there is no universally optimal way of incorporating SFMs, the complex prediction head gives the best performance for most tasks, although it increases the inference time. We also introduce an open-source toolkit and performance leaderboard, SLUE-PERB, for these tasks and modeling strategies.

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。

↑