dataset_info.json

{
  "name": "Pashto Synthetic Speech Dataset (2k)",
  "version": "1.0.0",
  "description": "Synthetic Pashto speech dataset with 2000 sentences in male and female voices",
  "license": "MIT",
  "created_at": "2025-03-13T13:16:53.610654",
  "statistics": {
    "total_sentences": 2000,
    "total_male_recordings": 2000,
    "total_female_recordings": 2000,
    "total_recordings": 4000,
    "sampling_rate": 44100,
    "total_male_duration_seconds": 0.0,
    "total_female_duration_seconds": 0.0,
    "average_male_duration_seconds": 0.0,
    "average_female_duration_seconds": 0.0,
    "longest_male_clip_seconds": 0.0,
    "longest_female_clip_seconds": 0.0,
    "shortest_male_clip_seconds": 0.0,
    "shortest_female_clip_seconds": 0.0
  },
  "voices": {
    "male": "ps-AF-GulNawazNeural",
    "female": "ps-AF-LatifaNeural"
  },
  "format": "WAV, 44.1kHz, 16-bit PCM"
}