classla/mak_na_konac
收藏Hugging Face2025-09-22 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/classla/mak_na_konac
下载链接
链接失效反馈官方服务:
资源简介:
Mak na Konac是一个克罗地亚语和塞尔维亚语的多参考语音识别基准数据集,包含大约20小时的自发口语样本转录。数据集目前包含音频和主键,但不包括多参考的金标准转录。
The Mak na Konac dataset is a multi-reference speech-to-text benchmark for Croatian and Serbian, consisting of transcribed spontaneous speech samples with a total duration of about 20 hours. Currently, the dataset includes audio and primary keys, but not the multi-reference gold truth transcriptions.
提供机构:
classla



