README.md

metadata

license: cc-by-sa-4.0
task_categories:
  - automatic-speech-recognition
language:
  - pl
pretty_name: PWr AZON
size_categories:
  - 10K<n<100K
dataset_info:
  features:
    - name: audio
      dtype:
        audio:
          sampling_rate: 44100
    - name: transcript
      dtype: string
    - name: gender
      dtype: string
    - name: id
      dtype: string
    - name: id_og
      dtype: string
  splits:
    - name: train
      num_bytes: 8585221408.406
      num_examples: 14491
    - name: unsup
      num_bytes: 1128648882
      num_examples: 841
  download_size: 9746452069
  dataset_size: 9713870290.406
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: unsup
        path: data/unsup-*

This speech dataset consists of 15332 short audio clips of multiple speakers speaking in Polish. Transcription is provided for 14491 audio clips (train split), and it is missing for 841 audio clips (unsup split). Gender of speaker is provided for the whole dataset. Clips have total length of almost 31 hours.

This dataset was created from Korpus nagrań próbek mowy do celów budowy modeli akustycznych dla automatycznego rozpoznawania mowy w języku polskim. The dataset was repackaged into easier to use format. If you are interested in the original data, please visit https://zasobynauki.pl/zasoby/korpus-nagran-probek-mowy-do-celow-budowy-modeli-akustycznych-dla-automatycznego-rozpoznawania-mowy,53293/

Also, if you find this resource helpful, kindly consider leaving a like.