feat: add JSON schema for scene training data (#647)

schemas/scene-description.schema.json

@@ -0,0 +1,104 @@
+{
+  "$schema": "http://json-schema.org/draft-07/schema#",
+  "title": "Scene Description Training Entry",
+  "description": "Schema for scene-descriptions-pop.jsonl training data entries",
+  "type": "object",
+  "required": [
+    "song",
+    "artist",
+    "genre",
+    "bpm",
+    "beat",
+    "timestamp",
+    "duration_seconds",
+    "lyric_line",
+    "scene"
+  ],
+  "properties": {
+    "song": {
+      "type": "string",
+      "minLength": 1,
+      "description": "Song title"
+    },
+    "artist": {
+      "type": "string",
+      "minLength": 1,
+      "description": "Artist name"
+    },
+    "genre": {
+      "type": "string",
+      "minLength": 1,
+      "description": "Music genre"
+    },
+    "bpm": {
+      "type": "integer",
+      "minimum": 20,
+      "maximum": 300,
+      "description": "Beats per minute"
+    },
+    "beat": {
+      "type": "integer",
+      "minimum": 1,
+      "description": "Beat number in sequence"
+    },
+    "timestamp": {
+      "type": "string",
+      "pattern": "^\\d+:\\d{2}$",
+      "description": "Timestamp as M:SS or MM:SS"
+    },
+    "duration_seconds": {
+      "type": "number",
+      "minimum": 0.1,
+      "maximum": 30,
+      "description": "Duration of this segment"
+    },
+    "lyric_line": {
+      "type": "string",
+      "minLength": 1,
+      "description": "Lyric text for this beat"
+    },
+    "scene": {
+      "type": "object",
+      "required": [
+        "mood",
+        "colors",
+        "composition",
+        "camera",
+        "description"
+      ],
+      "properties": {
+        "mood": {
+          "type": "string",
+          "minLength": 1,
+          "description": "Emotional mood"
+        },
+        "colors": {
+          "type": "array",
+          "items": {
+            "type": "string",
+            "minLength": 1
+          },
+          "minItems": 1,
+          "description": "Color palette"
+        },
+        "composition": {
+          "type": "string",
+          "minLength": 1,
+          "description": "Shot composition"
+        },
+        "camera": {
+          "type": "string",
+          "minLength": 1,
+          "description": "Camera movement"
+        },
+        "description": {
+          "type": "string",
+          "minLength": 10,
+          "description": "Scene description"
+        }
+      },
+      "additionalProperties": false
+    }
+  },
+  "additionalProperties": false
+}