MERaLiON/Multitask-National-Speech-Corpus-v1
收藏Hugging Face2025-01-21 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/MERaLiON/Multitask-National-Speech-Corpus-v1
下载链接
链接失效反馈官方服务：
资源简介：
多任务国家语音语料库（MNSC v1）是从IMDA的NSC语料库派生出来的。这是一个针对语音理解的多任务语料库，进一步注释了新加坡本地口音、本地化术语和代码转换的知识。它包括自动语音识别（ASR）、语音问答（SQA）、口语对话摘要（SDS）和副语言问答（PQA）任务。
The Multitask-National-Speech-Corpus (MNSC v1) is derived from IMDAs NSC Corpus. It is a multitask speech understanding dataset that has been further annotated with knowledge of Singapores local accent, localized terms, and code-switching. It includes tasks such as Automatic Speech Recognition (ASR), Speech Question Answering (SQA), Spoken Dialogue Summarization (SDS), and Paralinguistic Question Answering (PQA).
提供机构：
MERaLiON
