---
dataset_info:
  features:
  - name: audio
    dtype:
      audio:
        sampling_rate: 16000
  - name: sentence
    dtype: string
  splits:
  - name: train
    num_bytes: 13794178060.666706
    num_examples: 72618
  - name: test
    num_bytes: 726198619.7152941
    num_examples: 3823
  download_size: 13879409095
  dataset_size: 14520376680.382
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
language:
- ja
pretty_name: 3 japanese audio datasets - proccessed - common voice 16.1, google fleurs, animespeech
license: apache-2.0
task_categories:
- automatic-speech-recognition
- translation
- audio-to-audio
- audio-classification
---