README.md

metadata

dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: text
      dtype: string
    - name: info
      dtype: string
  splits:
    - name: train
      num_bytes: 92594977
      num_examples: 419
    - name: validation
      num_bytes: 7098201
      num_examples: 27
    - name: test
      num_bytes: 27159782
      num_examples: 113
  download_size: 125723439
  dataset_size: 126852960
language:
  - en
tags:
  - air traffic management
  - natural language processing
  - ATCO2
  - automatic speech recognition
  - atm
  - asr
pretty_name: ATCO2-ASR Data

Dataset Card for "ATCO2-ASR"

This is audio data used for automatic speech recognition. The original source of the data is the ATCO2 project, specifically the ASR part of the public speech corpus.