KTH/nst

---
license: cc0-1.0
task_categories:
- automatic-speech-recognition
language:
- sv
---
# NST Swedish ASR Database (16 kHz) – reorganized

This database was created by Nordic Language Technology for the development of automatic speech recognition and dictation in Swedish. In this updated version, the organization of the data have been altered to improve the usefulness of the database.

In the original version of the material, the files were organized in a specific folder structure where the folder names were meaningful. However, the file names were not meaningful, and there were also cases of files with identical names in different folders. This proved to be impractical, since users had to keep the original folder structure in order to use the data. The files have been renamed, such that the file names are unique and meaningful regardless of the folder structure. The original metadata files were in spl format. These have been converted to JSON format. The converted metadata files are also anonymized and the text encoding has been converted from ANSI to UTF-8.

See the documentation file for a full description of the data and the changes made to the database.

The data is originally hosted on the National Library of Norway website.
https://www.nb.no/sprakbanken/en/resource-catalogue/oai-nb-no-sbr-56/

Hosting on Hugging Face datasets for convenience.

Licence
CC0 1.0 Universal (CC0 1.0)
Public Domain Dedication

原始信息汇总

数据集概述

数据集名称

NST Swedish ASR Database (16 kHz) – reorganized

创建机构

Nordic Language Technology

数据集用途

用于开发瑞典语的自动语音识别和听写系统。

数据集更新内容

• 文件命名已更新，确保文件名唯一且有意义，不再依赖原始文件夹结构。
• 原始的spl格式元数据文件已转换为JSON格式，并进行了匿名化处理。
• 文本编码从ANSI转换为UTF-8。

数据集语言

瑞典语（sv）

任务类别

• 自动语音识别

许可证

CC0 1.0 Universal (CC0 1.0) 公共领域贡献