arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2607.03201eess.AScs.LGcs.SD
收藏

从长时长录音推导基准数据集:挑战与机遇

Deriving Benchmarking Datasets from Long-Form Recordings: Challenges and Opportunities

LAAC, LSCP, DEC, ENS, EHESS, CNRS, PSL University, Paris, France(法国巴黎PSL大学、CNRS、EHESS、ENS、DEC、LSCP、LAAC);Laboratoire d'Informatique et Systèmes, Université Aix-Marseille, CNRS, France(法国国家科学研究中心、艾克斯-马赛大学信息与系统实验室);Signal Processing Research Centre, Tampere University, Finland(芬兰塔尔库大学信号处理研究中心)

Kaveri K. Sheth, Lawrence Borst, Tarek Kunze, Marvin Lavechin, Okko Räsänen, Sho Tsuji, Loann Peurey, Alix Bourrée, Alejandrina Cristia

AI总结:

针对儿童长时长录音用于早期语言研究的跨库复用难、缺标准基准、隐私合规差问题,提出含标准化数据集、可复现流水线、伦理生态系统的综合框架。

摘要:

以儿童为核心的长时长录音(LFR)是研究早期语言发展的生态有效性来源,但三类问题限制了其应用:第一,LFR语料库在不同站点采集,格式与授权结构异质,跨库复用难度大;第二,缺乏标准化基准,难以评估工具在不同语言与场景下的泛化能力;第三,机器学习工作流通常无法遵守敏感儿童语音对应的隐私约束。本文提出一套可同时解决上述三类问题的框架:基于开源工具构建的27个儿童导向数据集的标准化集合(S1)、面向四类语音处理基准的可复现流水线(S2),以及将伦理治理嵌入机器学习工作流的角色导向生态系统ELSI。本文通过语音类型分类案例研究验证该框架,证明三类解决方案相互依存。

英文摘要:

Long-form recordings (LFRs) of child-centered audio are ecologically valid sources for studying early language development, but three problems limit their use. First, LFR corpora are collected across sites with heterogeneous formats and consent structures, making cross-corpus use non-trivial. Second, without standardized benchmarks, assessing whether tools generalize across languages and conditions is hard. Third, ML workflows rarely respect privacy constraints governing sensitive child speech. This paper presents a framework addressing all three: a standardized collection of 27 child-centered datasets built with open-source tools (S1); a replicable pipeline for four speech-processing benchmarks (S2); and ELSI, a role-based ecosystem embedding ethical governance into the ML workflow (S3). We demonstrate the framework via a voice type classification case study and show the three solutions are mutually dependent.

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。

↑