Aditya-Sarvam's picture
Upload dataset
98dc53c verified
metadata
dataset_info:
  features:
    - name: audio_filepath
      dtype:
        audio:
          sampling_rate: 16000
    - name: language
      dtype:
        class_label:
          names:
            '0': bn-IN
            '1': hi-IN
            '2': kn-IN
            '3': ml-IN
            '4': ta-IN
            '5': gu-IN
            '6': mr-IN
            '7': te-IN
            '8': pa-IN
            '9': en-IN
            '10': or-IN
    - name: transcription
      dtype: string
    - name: translation
      dtype: string
    - name: duration
      dtype: float32
    - name: file_name
      dtype: string
  splits:
    - name: bengali
      num_bytes: 59574616
      num_examples: 315
    - name: english
      num_bytes: 43169725
      num_examples: 300
    - name: hindi
      num_bytes: 67533560
      num_examples: 337
    - name: kannada
      num_bytes: 60628042
      num_examples: 326
    - name: malayalam
      num_bytes: 53515163
      num_examples: 278
    - name: tamil
      num_bytes: 64895946
      num_examples: 328
    - name: gujarati
      num_bytes: 69209451
      num_examples: 362
    - name: marathi
      num_bytes: 54080600
      num_examples: 293
    - name: telugu
      num_bytes: 54159851
      num_examples: 277
    - name: assamese
      num_bytes: 19804671
      num_examples: 142
    - name: punjabi
      num_bytes: 47048351
      num_examples: 291
  download_size: 571644387
  dataset_size: 593619976
configs:
  - config_name: default
    data_files:
      - split: bengali
        path: data/bengali-*
      - split: english
        path: data/english-*
      - split: hindi
        path: data/hindi-*
      - split: kannada
        path: data/kannada-*
      - split: malayalam
        path: data/malayalam-*
      - split: tamil
        path: data/tamil-*
      - split: gujarati
        path: data/gujarati-*
      - split: marathi
        path: data/marathi-*
      - split: telugu
        path: data/telugu-*
      - split: assamese
        path: data/assamese-*
      - split: punjabi
        path: data/punjabi-*