Skip to content

Dataset · Jzuluaga

ATCO2 1-hour air traffic control test corpus

Jzuluaga/atco2_corpus_1h

A small audio test corpus of air traffic control radio communications with English transcriptions for ASR verification.

What's inside

A 1-hour test corpus containing under 1,000 recorded air traffic control voice communications with English text transcriptions. The data was collected and preprocessed under the European ATCO2 project with support from the Horizon 2020 program. The dataset targets automatic speech recognition (ASR) tasks in complex aviation environments.

What you can do with it

  • Testing ASR models (such as Whisper) on specific terminology and radio static noise
  • Evaluating speech recognition quality in voice control and monitoring systems for aviation
Tasks
Automatic speech recognition
Languages
English
Size
up to 1K rows
Downloads this month
788
Likes
13

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("Jzuluaga/atco2_corpus_1h")