Datasets:

mikelalda
/

basque_speech_dataset

configs:
  - config_name: default
    data_files:
      - split: emakumezkoa
        path: data/emakumezkoa-*
      - split: gizonezkoa
        path: data/gizonezkoa-*
dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: transcription
      dtype: string
  splits:
    - name: emakumezkoa
      num_bytes: 2487307163.632
      num_examples: 3858
    - name: gizonezkoa
      num_bytes: 2272963307.88
      num_examples: 3278
  download_size: 3798014109
  dataset_size: 4760270471.512

Basque Speech Dataset

This dataset contains Crowdsourced high-quality Basque speech data.

It includes audio samples and their corresponding transcriptions.

Splits:

emakumezkoa: Female speakers
gizonezkoa: Male speakers

Source: The data is sourced from OpenSLR - Dataset 76.

License: (Please refer to the LICENSE file in the dataset repository for detailed license information)