feat: implement full-platform STT API (v2.3 consensus plan)

Batch+realtime transcription API: faster-whisper engine w/ EngineOwner (single GPU owner, OOM downgrade chain, persisted attempted_profiles), hardware-adaptive device manager (T0-T3 VRAM tiers), Redis/in-proc job queue w/ leases + crash recovery, postprocess (glossary/rules/LLM w/ egress guard), privacy-first result store (UUID keys, source deleted after transcribe), retention sweeper, API-key auth (HMAC digests, scopes, job ownership), WebSocket realtime lane (LocalAgreement), CLI (detect/transcribe/bench/serve/key), Docker, benchmark runner. 127 mock-based tests pass; ruff clean. Includes verification checklist and autoplan review notes.
2026-08-12 16:01:21 +09:00
parent b7c30f8b71
commit 7327145d7a
82 changed files with 7351 additions and 0 deletions
@@ -0,0 +1,83 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "$id": "https://git.lukehemmin.com/lukehemmin/luke_scribe/docs/schemas/transcript-result-v1.schema.json",
+  "title": "TranscriptResult v1",
+  "type": "object",
+  "required": ["schema_version", "status"],
+  "properties": {
+    "schema_version": { "const": "1.0" },
+    "status": { "enum": ["completed", "failed", "cancelled"] },
+    "source": {
+      "type": "object",
+      "properties": {
+        "name": { "type": "string" },
+        "codec": { "type": ["string", "null"] },
+        "size_bytes": { "type": ["integer", "null"] }
+      }
+    },
+    "normalized_audio": {
+      "type": "object",
+      "properties": {
+        "duration_sec": { "type": "number", "exclusiveMinimum": 0 },
+        "audio_format": { "type": "string" },
+        "sample_rate": { "type": "integer" },
+        "channels": { "type": "integer" }
+      }
+    },
+    "execution": {
+      "type": "object",
+      "properties": {
+        "model": { "type": "string" },
+        "device": { "type": "string" },
+        "compute_type": { "type": "string" },
+        "language_requested": { "type": ["string", "null"] },
+        "language_detected": { "type": ["string", "null"] },
+        "language_detection_confidence": { "type": ["number", "null"] },
+        "model_used": { "type": ["string", "null"] },
+        "compute_type_used": { "type": ["string", "null"] },
+        "downgrade_attempts": { "type": "integer" },
+        "capability_tier": { "type": ["string", "null"] }
+      }
+    },
+    "timings": {
+      "type": "object",
+      "properties": {
+        "model_load_sec": { "type": ["number", "null"] },
+        "transcription_sec": { "type": ["number", "null"] },
+        "ingest_sec": { "type": ["number", "null"] },
+        "postprocess_sec": { "type": ["number", "null"] },
+        "rtf": { "type": ["number", "null"] }
+      }
+    },
+    "text": { "type": "string" },
+    "segments": {
+      "type": "array",
+      "items": {
+        "type": "object",
+        "required": ["index", "start", "end", "text"],
+        "properties": {
+          "index": { "type": "integer", "minimum": 0 },
+          "start": { "type": "number", "minimum": 0 },
+          "end": { "type": "number", "minimum": 0 },
+          "text": { "type": "string" },
+          "avg_logprob": { "type": ["number", "null"] },
+          "no_speech_prob": { "type": ["number", "null"] },
+          "confidence": { "type": ["number", "null"] },
+          "words": { "type": ["array", "null"] },
+          "speaker": { "type": ["string", "null"] }
+        }
+      }
+    },
+    "warnings": { "type": "array", "items": { "type": "string" } },
+    "error": {
+      "type": "object",
+      "properties": {
+        "code": { "type": "string" },
+        "message": { "type": "string" },
+        "retryable": { "type": "boolean" }
+      }
+    },
+    "postprocessing": { "type": ["object", "null"] },
+    "entities": { "type": ["array", "null"] }
+  }
+}