README.md

metadata

license: apache-2.0
task_categories:
  - automatic-speech-recognition
language:
  - bn
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*
      - split: eval
        path: data/eval-*
dataset_info:
  features:
    - name: transcriptions
      dtype: string
    - name: district
      dtype: string
    - name: input_features
      sequence:
        sequence: float32
    - name: labels
      sequence: int64
  splits:
    - name: train
      num_bytes: 11571309495.599985
      num_examples: 12007
    - name: eval
      num_bytes: 1286557689.4000149
      num_examples: 1335
  download_size: 7213173351
  dataset_size: 12857867185

This is the dataset of ভাষা-বিচিত্রা: ASR for Regional Dialects competition.

here i preprocessed and make train and eval split. this dataset consist of 10 dialact named 'barishal', 'chittagong', 'habiganj', 'kishoreganj', 'narail', 'narsingdi', 'rangpur', 'sandwip', 'sylhet', 'tangail'.

barishal district has 796 samples
chittagong district has 1406 samples
habiganj district has 940 samples
kishoreganj district has 1638 samples
narail district has 1488 samples
narsingdi district has 1098 samples
rangpur district has 1037 samples
sandwip district has 1049 samples
sylhet district has 2903 samples
tangail district has 987 samples