pathological_speech .json

{
  "description": "Held-out eval set - speakers NOT used in training/synthesis",
  "excluded_speakers": {
    "TORGO": [
      "F04",
      "FC02",
      "M02",
      "MC01"
    ],
    "UA-Speech": [
      "F02",
      "M04"
    ]
  },
  "included_speakers": [
    "F01",
    "F03",
    "F04",
    "FC01",
    "FC03",
    "M01",
    "M03",
    "M04",
    "M05",
    "M07",
    "M09",
    "M10",
    "MC02",
    "MC03",
    "MC04"
  ],
  "total_samples": 40241,
  "by_severity": {
    "Normal": 6636,
    "Moderate": 17635,
    "Severe": 12355,
    "Mild": 3615
  },
  "by_corpus": {
    "TORGO": 10765,
    "UA-Speech": 29476
  },
  "by_condition": {
    "Control": 6636,
    "Dysarthric": 33605
  },
  "by_speaker": {
    "FC01": 302,
    "FC03": 1922,
    "MC02": 1123,
    "MC03": 1669,
    "MC04": 1620,
    "M05": 5942,
    "M04": 666,
    "M01": 3548,
    "F01": 236,
    "M03": 810,
    "F03": 1087,
    "F04": 5251,
    "M09": 5355,
    "M10": 5355,
    "M07": 5355
  },
  "columns": [
    "audio",
    "text",
    "speaker_id",
    "corpus",
    "gender",
    "condition",
    "diagnosis",
    "severity",
    "intelligibility",
    "duration"
  ],
  "audio_sample_rate": 16000
}