Upload dataset_info.json with huggingface_hub

dataset_info.json

@@ -0,0 +1,60 @@
+{
+  "name": "Pashto Synthetic Speech Dataset Parquet (3k)",
+  "version": "1.0.0",
+  "description": "Synthetic Pashto speech dataset with 3000 sentences in male and female voices (Parquet format)",
+  "license": "MIT",
+  "created_at": "2025-03-21T11:35:51.374958",
+  "format": "parquet",
+  "splits": {
+    "train": {
+      "name": "train",
+      "num_examples": 6000
+    }
+  },
+  "statistics": {
+    "total_sentences": 3000,
+    "total_male_recordings": 3000,
+    "total_female_recordings": 3000,
+    "total_recordings": 6000,
+    "sampling_rate": 44100,
+    "total_male_duration_seconds": 1596.2710204081632,
+    "total_female_duration_seconds": 1635.4302040816326,
+    "average_male_duration_seconds": 0.5320903401360544,
+    "average_female_duration_seconds": 0.5451434013605442,
+    "longest_male_clip_seconds": 1.0906122448979592,
+    "longest_female_clip_seconds": 1.1248979591836734,
+    "shortest_male_clip_seconds": 0.12408163265306123,
+    "shortest_female_clip_seconds": 0.12408163265306123
+  },
+  "voices": {
+    "male": "ps-AF-GulNawazNeural",
+    "female": "ps-AF-LatifaNeural"
+  },
+  "audio_format": "WAV, 44.1kHz, 16-bit PCM",
+  "features": {
+    "sentence_id": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "text": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "audio": {
+      "_type": "Audio",
+      "sampling_rate": 44100
+    },
+    "duration": {
+      "dtype": "float64",
+      "_type": "Value"
+    },
+    "speaker": {
+      "dtype": "string",
+      "_type": "Value"
+    },
+    "sampling_rate": {
+      "dtype": "int64",
+      "_type": "Value"
+    }
+  }
+}