Hugging Face's logo

Datasets:
ivrit-ai
/
audio-transcripts

Tasks:
Audio Classification
Voice Activity Detection
Modalities:
Text
Formats:
parquet
Languages:
Hebrew
Size:
1M - 10M
ArXiv:
Libraries:
Datasets
Dask
License:
Dataset card Data Studio Files
xet
Community
5
audio-transcripts / README.md
yanirmr's picture
yanirmr
Update README.md
5c33e32 verified
preview code
|
raw
history blame contribute delete
2.85 kB
metadata
language:
  - he
license: other
size_categories:
  - 1M<n<10M
task_categories:
  - audio-classification
  - voice-activity-detection
license_name: ivrit.ai
license_link: LICENSE
extra_gated_fields:
  I have read the license, and agree to its terms: checkbox
dataset_info:
  features:
    - name: source
      dtype: string
    - name: episode
      dtype: string
    - name: uuid
      dtype: string
    - name: text
      dtype: string
    - name: attrs
      struct:
        - name: segments
          list:
            - name: avg_logprob
              dtype: float64
            - name: compression_ratio
              dtype: float64
            - name: end
              dtype: float64
            - name: id
              dtype: int64
            - name: no_speech_prob
              dtype: float64
            - name: seek
              dtype: int64
            - name: start
              dtype: float64
            - name: text
              dtype: string
  splits:
    - name: train
      num_bytes: 1290457176
      num_examples: 2183042
  download_size: 421521923
  dataset_size: 1290457176
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

ivrit.ai is a database of Hebrew audio and text content.

audio-base contains the raw, unprocessed sources.

audio-vad contains audio snippets generated by applying Silero VAD (https://github.com/snakers4/silero-vad) to the base dataset.

audio-transcripts contains transcriptions for each snippet in the audio-vad dataset.

The audio-base dataset contains data from the following sources:

Paper: https://arxiv.org/abs/2307.08720

If you use our datasets, the following quote is preferable:

@misc{marmor2023ivritai,
      title={ivrit.ai: A Comprehensive Dataset of Hebrew Speech for AI Research and Development}, 
      author={Yanir Marmor and Kinneret Misgav and Yair Lifshitz},
      year={2023},
      eprint={2307.08720},
      archivePrefix={arXiv},
      primaryClass={eess.AS}
}

License

The dataset is released under the ivrit.ai License, which enables broad research and commercial use.