metadata

dataset_info:
  features:
    - name: audio
      dtype: audio
    - name: transcription
      dtype: string
  splits:
    - name: train
      num_bytes: 1059254876.16
      num_examples: 2420
  download_size: 1055934511
  dataset_size: 1059254876.16
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
task_categories:
  - automatic-speech-recognition
size_categories:
  - 1K<n<10K
language:
  - xh

High quality TTS data for four South African languages - isiXhosa

Source - https://openslr.org/32/

Identifier: SLR32

Summary: Multi-speaker TTS data for four South African languages - isiXhosa

License: Attribution-ShareAlike 4.0 International (CC BY-SA 4.0)

About this resource:

This data set contains multi-speaker high quality transcribed audio data for four languages of South Africa. The data set consists of wave files, and a TSV file transcribing the audio. In each folder, the file line_index.tsv contains a FileID, which in turn contains the UserID and the Transcription of audio in the file. The data set has had some quality checks, but there might still be errors.

This data set was collected by as a collaboration between North West University and Google.

If you use this data in publications, please cite it as follows:

  @inproceedings{van-niekerk-etal-2017,
    title = {{Rapid development of TTS corpora for four South African languages}},
    author = {Daniel van Niekerk and Charl van Heerden and Marelie Davel and Neil Kleynhans and Oddur Kjartansson and Martin Jansche and Linne Ha},
    booktitle = {Proc. Interspeech 2017},
    pages = {2178--2182},
    address = {Stockholm, Sweden},
    month = aug,
    year  = {2017},
    URL   = {http://dx.doi.org/10.21437/Interspeech.2017-1139}
  }