UWB-ATCC corpus

该数据集由西波希米亚大学发布，名为UWB-ATCC语料库，它是一个面向航空交通管制（ATC）研究的免费公共资源，包含了航空交通管制员与飞行员之间的通信录音。这些语音资料经过了手动转录，并标注了说话者信息，即区分了飞行员和管制员。音频数据为单声道，采样率为8kHz，16位PCM格式。该数据集总时长约为13小时，适用于自动语音识别（ASR）和自然语言处理（NLP）任务。

Released by the University of West Bohemia, this dataset is named UWB-ATCC Corpus. It serves as a free public resource for air traffic control (ATC) research, comprising audio recordings of communications between air traffic controllers and pilots. All speech materials have undergone manual transcription and are annotated with speaker identities to differentiate between pilots and air traffic controllers. The audio data is in mono format, with a sampling rate of 8 kHz and 16-bit PCM encoding. The total duration of the dataset is approximately 13 hours, making it suitable for automatic speech recognition (ASR) and natural language processing (NLP) tasks.

搜集汇总

背景与挑战

背景概述

该数据集是一个包含20小时空中交通管制通信录音的语料库，语音经过人工转录并标注了说话者身份（飞行员/管制员），音频格式为8kHz、16bit PCM、单声道。数据集由西波西米亚大学发布，主要用于语音和声学模型研究。