CleanSky EC-H2020 ATCO2
收藏arXiv2020-08-13 更新2024-08-06 收录
下载链接：
http://arxiv.org/abs/2006.10304v2
下载链接
链接失效反馈官方服务：
资源简介：
CleanSky EC-H2020 ATCO2数据集由Idiap研究所创建，旨在解决航空交通控制（ATC）环境中语音通信的自动化问题。该数据集包含超过176小时的ATC语音数据，来源于多个国家和机场，涵盖多种口音和词汇变体。数据集的创建过程涉及从公开可访问的无线电频率频道收集数据，并进行自动预处理。该数据集主要用于训练和评估ASR系统，以提高其在ATC环境中的性能，特别是在处理非英语口音和环境噪音方面的能力。
The CleanSky EC-H2020 ATCO2 dataset was created by the Idiap Research Institute, aiming to address the automation of voice communications in air traffic control (ATC) environments. This dataset contains over 176 hours of ATC speech data sourced from multiple countries and airports, covering a wide range of accents and lexical variants. The dataset creation process involves collecting data from publicly accessible radio frequency channels and performing automated preprocessing. It is primarily employed for training and evaluating automatic speech recognition (ASR) systems to improve their performance in ATC environments, especially their ability to handle non-English accents and ambient environmental noise.
提供机构：
Idiap研究所
创建时间：
2020-06-18
