README.md

metadata

language:
  - pt
  - de
  - fr
  - sv
  - it
  - es
  - nl
license: mit
pretty_name: JWLang Corpus
datasets:
  - jwlang
tags:
  - automatic-speech-recognition
  - speech
  - dataset
  - jw.org
  - multilingual
  - whisper
viewer: true
dataset_info:
  - config_name: de
    features:
      - name: client_id
        dtype: string
      - name: audio
        dtype: audio
      - name: sentence
        dtype: string
      - name: language
        dtype: string
      - name: split
        dtype: string
    splits:
      - name: train
        num_bytes: 44420148
        num_examples: 949
      - name: test
        num_bytes: 5730879
        num_examples: 119
      - name: val
        num_bytes: 5849167
        num_examples: 119
    download_size: 223549840
    dataset_size: 56000194
  - config_name: es
    features:
      - name: client_id
        dtype: string
      - name: audio
        dtype: audio
      - name: sentence
        dtype: string
      - name: language
        dtype: string
      - name: split
        dtype: string
    splits:
      - name: train
        num_bytes: 43155769
        num_examples: 973
      - name: test
        num_bytes: 5317858
        num_examples: 122
      - name: val
        num_bytes: 5350687
        num_examples: 122
    download_size: 53591020
    dataset_size: 53824314
  - config_name: fr
    features:
      - name: client_id
        dtype: string
      - name: audio
        dtype: audio
      - name: sentence
        dtype: string
      - name: language
        dtype: string
      - name: split
        dtype: string
    splits:
      - name: train
        num_bytes: 40751557
        num_examples: 939
      - name: test
        num_bytes: 5126357
        num_examples: 118
      - name: val
        num_bytes: 5393533
        num_examples: 117
    download_size: 102271952
    dataset_size: 51271447
  - config_name: it
    features:
      - name: client_id
        dtype: string
      - name: audio
        dtype: audio
      - name: sentence
        dtype: string
      - name: language
        dtype: string
      - name: split
        dtype: string
    splits:
      - name: train
        num_bytes: 39862874
        num_examples: 779
      - name: test
        num_bytes: 5231142
        num_examples: 98
    download_size: 44992912
    dataset_size: 45094016
  - config_name: pt
    features:
      - name: client_id
        dtype: string
      - name: audio
        dtype: audio
      - name: sentence
        dtype: string
      - name: language
        dtype: string
      - name: split
        dtype: string
    splits:
      - name: train
        num_bytes: 45540940.152
        num_examples: 1004
      - name: test
        num_bytes: 5906213
        num_examples: 126
      - name: val
        num_bytes: 5474968
        num_examples: 125
    download_size: 340665914
    dataset_size: 56922121.152
configs:
  - config_name: de
    data_files:
      - split: train
        path: de/train-*
      - split: test
        path: de/test-*
      - split: val
        path: de/val-*
  - config_name: es
    data_files:
      - split: train
        path: es/train-*
      - split: test
        path: es/test-*
      - split: val
        path: es/val-*
  - config_name: fr
    data_files:
      - split: train
        path: fr/train-*
      - split: test
        path: fr/test-*
      - split: val
        path: fr/val-*
  - config_name: it
    data_files:
      - split: train
        path: it/train-*
      - split: test
        path: it/test-*
  - config_name: pt
    data_files:
      - split: train
        path: pt/train-*
      - split: test
        path: pt/test-*
      - split: val
        path: pt/val-*

JWLang Corpus

Dataset Summary

The JWLang Corpus is a collection of audio and corresponding text data from JW Broadcasting videos available on the jw.org website. It is designed for training and fine-tuning automatic speech recognition (ASR) models, specifically OpenAI Whisper. The dataset is stored in Parquet format on Hugging Face, with original audio files in MP3 format and corresponding text files. The data were downloaded in June 2024.

Splits

• Train
• Validation
• Test

Usage

To load and use the dataset:

from datasets import load_dataset

dataset = load_dataset("M2LabOrg/jwlang")

Example Data

Example text snippet from the dataset:

{
  "audio": "path/to/audio.mp3",
  "text": "Example subtitle text."
}

License

This dataset is private and intended for internal use only.

Citation

If you use this dataset, please cite:

@article{jwlang_corpus,
  title={JWLang Corpus from jw.org Videos for ASR Training},
  author={Michel Mesquita},
  journal={Unpublished},
  year={2024},
  note={Data downloaded from jw.org in June 2024 and processed by M. Mesquita}
}

Contact

For any questions or issues, please contact Michel Mesquita.