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 2307.01715cs.CLcs.LGcs.SDeess.AS
收藏

Align With Purpose:采用通用即插即用框架优化CTC模型的期望属性

Align With Purpose: Optimize Desired Properties in CTC Models with a General Plug-and-Play Framework

Orcam

Eliya Segev, Maya Alroy, Ronen Katsir, Noam Wies, Ayana Shenhav, Yael Ben-Oren, David Zar, Oren Tadmor, Jacob Bitterman, Amnon Shashua, Tal Rosenwein

AI总结:

针对CTC准则仅二元区分对齐、无法捕捉关键属性的问题,提出即插即用框架Align With Purpose,通过补充损失项按期望属性优化对齐,在ASR任务的延迟和WER优化上均取得显著效果,且易扩展。

摘要:

连接主义时序分类(Connectionist Temporal Classification,CTC)是训练监督序列到序列(seq2seq)模型的广泛使用的准则。它通过对生成真实标签的完美对齐进行边际化、以牺牲不完美对齐为代价,实现输入与输出序列之间的关系(即对齐)的学习。这种对完美和不完美对齐的二元区分,无法捕捉在其他现实应用中具有重要意义的其他关键对齐属性。\n本文提出了$\textit{Align With Purpose}$,这是一种$\textbf{通用即插即用框架}$,用于增强采用CTC准则训练的模型的期望属性。我们通过为CTC补充一个额外的损失项来实现这一点,该损失项会根据期望属性对对齐进行优先级排序。我们的方法无需对CTC损失函数进行任何干预,能够轻松优化多种属性,并且可以区分完美对齐与不完美对齐。\n我们将该框架应用于自动语音识别(Automatic Speech Recognition,ASR)领域,并从属性选择、架构选择以及训练数据集规模(最高达280000小时)三个维度证明了其通用性。为验证框架的有效性,我们将其应用于两个不相关的属性:发射时间与词错误率(word error rate,WER)。对于前者,我们报告在延迟优化方面实现了最高570ms的提升,同时WER仅有小幅下降;对于后者,我们报告相较于基线模型,WER实现了4.5%的相对提升。据我们所知,此类应用从未在如此大规模的数据集上被验证可行。值得注意的是,我们的方法仅需几行代码即可实现,并且可以扩展到其他无对齐损失函数以及ASR以外的领域。

英文摘要:

Connectionist Temporal Classification (CTC) is a widely used criterion for training supervised sequence-to-sequence (seq2seq) models. It enables learning the relations between input and output sequences, termed alignments, by marginalizing over perfect alignments (that yield the ground truth), at the expense of imperfect alignments. This binary differentiation of perfect and imperfect alignments falls short of capturing other essential alignment properties that hold significance in other real-world applications. Here we propose $\textit{Align With Purpose}$, a $\textbf{general Plug-and-Play framework}$ for enhancing a desired property in models trained with the CTC criterion. We do that by complementing the CTC with an additional loss term that prioritizes alignments according to a desired property. Our method does not require any intervention in the CTC loss function, enables easy optimization of a variety of properties, and allows differentiation between both perfect and imperfect alignments. We apply our framework in the domain of Automatic Speech Recognition (ASR) and show its generality in terms of property selection, architectural choice, and scale of training dataset (up to 280,000 hours). To demonstrate the effectiveness of our framework, we apply it to two unrelated properties: emission time and word error rate (WER). For the former, we report an improvement of up to 570ms in latency optimization with a minor reduction in WER, and for the latter, we report a relative improvement of 4.5% WER over the baseline models. To the best of our knowledge, these applications have never been demonstrated to work on a scale of data as large as ours. Notably, our method can be implemented using only a few lines of code, and can be extended to other alignment-free loss functions and to domains other than ASR.

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。

↑