---
dataset_info:
  features:
  - name: speaker_id
    dtype: string
  - name: gender
    dtype: string
  - name: transcript
    dtype: string
  - name: interval
    sequence: float64
  - name: untranscripted
    dtype: bool
  - name: topic
    dtype: string
  - name: audio
    dtype: audio
  splits:
  - name: train
    num_bytes: 234737425.48921812
    num_examples: 3078
  - name: test
    num_bytes: 12430884.226781858
    num_examples: 163
  download_size: 242347299
  dataset_size: 247168309.71599996
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---


Data set source: https://magichub.com/datasets/egyptian-arabic-conversational-speech-corpus/