Updates to intel and correlation

drb-c2-core/app/config.py

@@ -24,7 +24,8 @@ class Settings(BaseSettings):
     gemini_api_key: Optional[str] = None
     summary_interval_minutes: int = 2          # how often the summary loop runs
     correlation_window_hours: int = 2          # slow/location path: max hours since last call
-    embedding_similarity_threshold: float = 0.93  # slow-path cosine threshold (tiebreaker only)
+    embedding_similarity_threshold: float = 0.93  # slow-path: requires location corroboration
+    embedding_no_location_threshold: float = 0.97 # slow-path: match without location (very high bar)
     location_proximity_km: float = 0.5        # radius for location-proximity matching
     incident_auto_resolve_minutes: int = 90   # auto-resolve after N minutes with no new calls
     recorrelation_scan_minutes: int = 15      # re-examine orphaned calls ended within this window

drb-c2-core/app/internal/incident_correlator.py

@@ -79,7 +79,25 @@ async def correlate_call(
 
     matched_incident: Optional[dict] = None
 
+    # A "thin" call carries no scene-identifying information — it is a pure
+    # status transmission (10-4, en route, acknowledgement).  Detected by the
+    # absence of extracted units, vehicles, AND geocoded coordinates.  Thin
+    # calls should link to wherever the last active conversation on this TGID
+    # was happening rather than running the full scene-verification logic.
+    is_thin_call = not call_units and not call_vehicles and not coords
+
     # ── 1. Fast path: talkgroup match (any type, no time limit) ──────────────
+    #
+    # Two distinct behaviours depending on call substance:
+    #
+    # • Thin call  → link to the most-recently-updated active incident on this
+    #               TGID (i.e. the last conversation in progress).
+    #
+    # • Substantive call (has location / units / vehicles) → verify the call
+    #   actually belongs to the matched incident before linking.  When a busy
+    #   dispatch channel runs multiple concurrent scenes (different addresses,
+    #   different units) we split them into separate incidents rather than
+    #   merging everything because they share a talkgroup.
     if talkgroup_id is not None and system_id:
         tg_str     = str(talkgroup_id)
         tg_matches = [
@@ -87,11 +105,30 @@ async def correlate_call(
             if system_id in (inc.get("system_ids") or [])
             and tg_str in (inc.get("talkgroup_ids") or [])
         ]
-        if len(tg_matches) == 1:
-            matched_incident = tg_matches[0]
+
+        if tg_matches and is_thin_call:
+            # Status/ack call — no scene data to reason about.
+            # Attach to whichever incident was most recently active on this TGID.
+            matched_incident = max(tg_matches, key=lambda inc: inc.get("updated_at", ""))
             logger.info(
-                f"Correlator fast-path: call {call_id} → {tg_matches[0]['incident_id']}"
+                f"Correlator fast-path (thin→last TGID incident): "
+                f"call {call_id} → {matched_incident['incident_id']}"
             )
+        elif len(tg_matches) == 1:
+            candidate = tg_matches[0]
+            if _call_fits_incident(
+                candidate, call_units, call_vehicles, coords, settings.location_proximity_km
+            ):
+                matched_incident = candidate
+                logger.info(
+                    f"Correlator fast-path: call {call_id} → {candidate['incident_id']}"
+                )
+            else:
+                logger.info(
+                    f"Correlator fast-path skipped: call {call_id} — different scene "
+                    f"from {candidate['incident_id']} (no unit overlap + distant location); "
+                    f"will attempt new incident"
+                )
         elif len(tg_matches) > 1:
             matched_incident = _disambiguate(
                 tg_matches, call_units, call_vehicles, coords, call_embedding
@@ -119,7 +156,13 @@ async def correlate_call(
                 )
                 break
 
-    # ── 3. Slow path: embedding + location corroboration (time-limited, same type) ──
+    # ── 3. Slow path: embedding similarity (time-limited, same type) ──────────
+    #
+    # Two tiers:
+    #   ① embedding_similarity_threshold + location corroboration (standard)
+    #   ② embedding_no_location_threshold alone — when geocoding failed on
+    #      either side but the transcript content is semantically very close.
+    #      A strong embedding match beats a missing geocode.
     if not matched_incident and call_embedding and incident_type:
         best_score = 0.0
         best_inc: Optional[dict] = None
@@ -147,7 +190,14 @@ async def correlate_call(
                         f"Correlator slow-path: call {call_id} → {best_inc['incident_id']} "
                         f"(sim={best_score:.3f}, dist={dist_km:.2f}km)"
                     )
-            # No coords available → slow path alone is not enough; skip
+            elif best_score >= settings.embedding_no_location_threshold:
+                # High-confidence semantic match; geocode unavailable on one or
+                # both sides — content similarity alone is sufficient evidence.
+                matched_incident = best_inc
+                logger.info(
+                    f"Correlator slow-path (high-confidence, no location): "
+                    f"call {call_id} → {best_inc['incident_id']} (sim={best_score:.3f})"
+                )
 
     # ── Update existing or create new ────────────────────────────────────────
     if matched_incident:
@@ -167,7 +217,6 @@ async def correlate_call(
         # No match and either no type or creation suppressed — nothing to do
         return None
 
-    await fstore.doc_set("calls", call_id, {"incident_id": incident_id})
     return incident_id
 
 
@@ -254,6 +303,62 @@ def _disambiguate(
     return best
 
 
+def _call_fits_incident(
+    inc: dict,
+    call_units: list[str],
+    call_vehicles: list[str],
+    call_coords: Optional[dict],
+    proximity_km: float,
+) -> bool:
+    """
+    Return True if this call plausibly belongs to the given incident.
+
+    This guards the single-talkgroup-match fast path on busy dispatch channels
+    where multiple concurrent scenes share one talkgroup.  We only return False
+    (→ create a new incident) when there is *positive evidence* of a different
+    scene: a geocoded location that is too far away AND no unit or vehicle
+    overlap.  In all ambiguous cases we default to True (link) to avoid
+    fragmenting short status calls that carry no location or unit information.
+
+    Examples that correctly split:
+      - Police dispatch sends units to two separate MVAs miles apart
+      - EMS handles overlapping aided cases at different addresses
+
+    Examples that correctly stay together (domestic with split parties):
+      - Units at 10 Main St and 12 Main St — within proximity radius → True
+      - Same unit mentioned in both the call and the incident → True
+    """
+    # Unit overlap is the strongest positive signal: same officers = same call.
+    inc_units = set(inc.get("units") or [])
+    if inc_units and call_units and any(u in inc_units for u in call_units):
+        return True
+
+    # Vehicle overlap: same vehicle description across calls → same scene.
+    inc_vehicles = set(inc.get("vehicles") or [])
+    if inc_vehicles and call_vehicles and any(v in inc_vehicles for v in call_vehicles):
+        return True
+
+    # When both sides have geocoded coordinates, distance is the tiebreaker.
+    inc_coords = inc.get("location_coords")
+    if call_coords and inc_coords:
+        dist_km = _haversine_km(
+            call_coords["lat"], call_coords["lng"],
+            inc_coords["lat"], inc_coords["lng"],
+        )
+        # Within proximity radius → same scene (handles domestics with nearby split parties).
+        if dist_km <= proximity_km:
+            return True
+        # Different location AND no unit/vehicle overlap → different incident.
+        return False
+
+    # No geocoded location on one or both sides but the call IS substantive
+    # (has units or vehicles — otherwise is_thin_call would have caught it).
+    # Unit overlap already returned True above if present.  If we reach here
+    # there is no overlap and no coords to compare — we cannot prove it is a
+    # different scene, so default to linking rather than fragmenting.
+    return True
+
+
 async def _update_incident(
     inc: dict,
     call_id: str,
@@ -313,6 +418,11 @@ async def _update_incident(
     if best_coords:
         updates["location_coords"] = best_coords
 
+    # Update incident type when a re-classified call provides a concrete type.
+    # This handles the case where admin correction changes fire→police, etc.
+    if incident_type and incident_type != inc.get("type"):
+        updates["type"] = incident_type
+
     # Re-evaluate title when a substantive call (classified incident_type) brings new tags.
     # Routine status calls (type=None) do not clobber the title.
     if incident_type:

drb-c2-core/app/internal/intelligence.py

@@ -1,8 +1,10 @@
 """
 GPT-4o-mini intelligence extraction from call transcripts.
 
-Sends the transcript to GPT-4o mini with a tight JSON schema prompt.
-Returns structured data: incident type, tags, location, vehicles, units, severity.
+Sends the transcript to GPT-4o-mini with a structured prompt that detects
+whether the recording contains one or multiple distinct scenes (back-to-back
+dispatch conversations on a busy channel).  Returns a list of scene dicts —
+one per detected incident.  Most calls produce a single scene.
 
 Falls back gracefully if the API is unavailable or returns malformed output.
 """
@@ -13,30 +15,37 @@ from typing import Optional
 from app.internal.logger import logger
 from app.internal import firestore as fstore
 
-_PROMPT_TEMPLATE = """You are analyzing a P25 public safety radio recording. The audio was transcribed by Whisper through a digital radio vocoder, which introduces errors. Each numbered transmission is a separate PTT press from a different radio. Extract structured information and respond ONLY with a single valid JSON object — no markdown, no explanation.
+_PROMPT_TEMPLATE = """You are analyzing a P25 public safety radio recording. The audio was transcribed by Whisper through a digital radio vocoder, which introduces errors. Each numbered transmission is a separate PTT press from a different radio.
 
-Schema:
-{{
-  "incident_type": one of "fire" | "ems" | "police" | "accident" | "other" | "unknown",
-  "tags": [list of specific descriptive tags, max 6, e.g. "two-car mva", "property-damage-only", "working fire", "shots-fired"],
-  "location": "most specific location string found, or empty string",
-  "vehicles": [vehicle descriptions mentioned, e.g. "Hyundai Tucson", "black sedan"],
-  "units": [unit IDs or officer numbers mentioned, e.g. "Unit 511", "Car 4"],
-  "severity": one of "minor" | "moderate" | "major" | "unknown",
-  "resolved": true if this call explicitly signals the incident is over ("Code 4", "in custody", "all clear", "fire out", "patient transported", "GOA", "scene clear", "10-42", "negative contact", "clear the scene"), false otherwise,
-  "transcript_corrected": "corrected full transcript string, or null if no corrections needed"
-}}
+SCENE DETECTION:
+A busy dispatch channel sometimes captures back-to-back conversations about multiple concurrent incidents in a single recording. Detect whether this recording contains ONE scene (all transmissions relate to a single event) or MULTIPLE scenes (clearly distinct dispatch conversations with different units being assigned, different locations, different event types). Assign short status transmissions (10-4, en route, acknowledgements) with no clear scene context to the most recent scene before them in the list.
+
+Always respond with the scenes array, even for a single scene.
+
+Response format — a JSON object with a "scenes" array. Each scene:
+  segment_indices: list of 0-based indices into the numbered transmissions (or null if no segments)
+  incident_type: one of "fire" | "ems" | "police" | "accident" | "other" | "unknown"
+  tags: list of specific descriptive tags, max 6, e.g. "two-car mva", "working fire", "shots-fired"
+  location: most specific location string found, or empty string
+  vehicles: list of vehicle descriptions mentioned
+  units: list of unit IDs or officer numbers explicitly mentioned
+  severity: one of "minor" | "moderate" | "major" | "unknown"
+  resolved: true if this scene explicitly signals incident closure, false otherwise
+  transcript_corrected: corrected text for this scene's transmissions only, or null
 
 Rules:
 - location: prefer intersections > addresses > mile markers > route+town > route alone > town alone. Empty string if none.
-- tags: be specific and lowercase, hyphenated. Do not repeat incident_type as a tag.
+- tags: specific, lowercase, hyphenated. Do not repeat incident_type as a tag.
 - units: only identifiers explicitly mentioned, not inferred.
 - Do not invent details not present in the transcript.
-- transcript_corrected: fix only clear STT errors caused by vocoder distortion (e.g. "Several" → "10-4", misheard street names, garbled unit IDs). Use the back-and-forth context between transmissions to resolve ambiguities. Keep all radio language as-is — do NOT decode codes into plain English. Return null if the transcript looks accurate.
+- incident_type: let the talkgroup channel be your primary signal. Use "fire" ONLY if the talkgroup is clearly a fire/rescue channel OR the transcript explicitly describes active fire, smoke, flames, or structure fire activation. Police or EMS referencing a fire scene → use "police" or "ems". When uncertain, prefer "other" over "fire".
+- ten_codes: interpret radio codes using the department reference provided below. Do not guess codes not listed.
+- resolved: true only when the scene explicitly signals "Code 4", "all clear", "10-42", "in custody", "patient transported", "fire out", "GOA", "negative contact", "scene clear".
+- transcript_corrected: fix only clear STT/vocoder errors (e.g. "Several" → "10-4", misheard street names, garbled unit IDs). Keep all radio language as-is — do NOT decode codes into plain English. Return null if accurate.
 
 System: {system_id}
 Talkgroup: {talkgroup_name}
-{vocabulary_block}{transcript_block}"""
+{ten_codes_block}{vocabulary_block}{transcript_block}"""
 
 # Nominatim viewbox half-width in degrees (~11 km at mid-latitudes)
 _GEO_DELTA = 0.1
@@ -54,7 +63,14 @@ _TG_SUFFIX_RE = re.compile(
 )
 
 
-async def extract_tags(
+def _build_ten_codes_block(ten_codes: dict[str, str]) -> str:
+    if not ten_codes:
+        return ""
+    lines = "\n".join(f"  {code}: {meaning}" for code, meaning in sorted(ten_codes.items()))
+    return f"Department ten-codes:\n{lines}\n\n"
+
+
+async def extract_scenes(
     call_id: str,
     transcript: str,
     talkgroup_name: Optional[str] = None,
@@ -63,84 +79,128 @@ async def extract_tags(
     segments: Optional[list[dict]] = None,
     node_id: Optional[str] = None,
     preserve_transcript_correction: bool = False,
-) -> tuple[list[str], Optional[str], Optional[str], Optional[dict], bool]:
+) -> list[dict]:
     """
-    Extract incident tags, type, location, corrected transcript, and closure signal via GPT-4o mini.
-    Geocodes the extracted location string via Nominatim using the node's position as bias.
+    Split the transcript into one or more scenes and extract structured
+    intelligence for each.  Most calls return a single scene; a busy dispatch
+    channel capturing back-to-back conversations returns multiple.
 
-    Returns:
-        (tags, primary_type, location_str, location_coords, resolved)
-        where location_coords is {"lat": float, "lng": float} or None,
-        and resolved is True when the transcript signals incident closure.
+    Each scene dict contains:
+        tags, incident_type, location, location_coords, resolved,
+        severity, vehicles, units, transcript_corrected,
+        segment_indices, embedding
 
-    Side-effect: updates calls/{call_id} in Firestore with tags, location,
-    location_coords, vehicles, units, severity, transcript_corrected; also stores embedding.
+    Side-effect: updates calls/{call_id} in Firestore with merged tags,
+    location (primary scene), units/vehicles, severity, embedding, and
+    optionally transcript_corrected.
     """
-    # Load per-system vocabulary for prompt injection
     vocabulary: list[str] = []
+    ten_codes: dict[str, str] = {}
     if system_id:
         from app.internal.vocabulary_learner import get_vocabulary
         vocab_data = await get_vocabulary(system_id)
         vocabulary = vocab_data.get("vocabulary") or []
+        system_doc = await fstore.doc_get("systems", system_id)
+        if system_doc:
+            ten_codes = system_doc.get("ten_codes") or {}
 
-    result = await asyncio.to_thread(
-        _sync_extract, transcript, talkgroup_name, talkgroup_id, system_id, segments, vocabulary
+    raw_scenes: list[dict] = await asyncio.to_thread(
+        _sync_extract,
+        transcript, talkgroup_name, talkgroup_id, system_id, segments, vocabulary, ten_codes,
     )
 
-    tags: list[str]              = result.get("tags") or []
-    incident_type: Optional[str] = result.get("incident_type") or None
-    location: Optional[str]      = result.get("location") or None
-    vehicles: list[str]          = result.get("vehicles") or []
-    units: list[str]             = result.get("units") or []
-    severity: str                = result.get("severity") or "unknown"
-    resolved: bool               = bool(result.get("resolved", False))
-    transcript_corrected: Optional[str] = result.get("transcript_corrected") or None
+    if not raw_scenes:
+        return []
 
-    if incident_type in ("unknown", "other", ""):
-        incident_type = None
-
-    # Geocode the location string if we have one and a node to bias toward
-    location_coords: Optional[dict] = None
-    if location and node_id:
+    # Resolve node position once for geocoding all scenes
+    node_lat: Optional[float] = None
+    node_lon: Optional[float] = None
+    if node_id:
         node_doc = await fstore.doc_get("nodes", node_id)
         if node_doc:
             node_lat = node_doc.get("lat")
             node_lon = node_doc.get("lon")
-            if node_lat is not None and node_lon is not None:
-                state      = await _get_node_state(node_id, node_lat, node_lon)
-                muni       = _municipality_from_tg(talkgroup_name)
-                hint_parts = [p for p in [muni, state] if p]
-                query      = f"{location}, {', '.join(hint_parts)}" if hint_parts else location
-                location_coords = await _geocode_location(query, node_lat, node_lon)
 
-    # Store embedding alongside structured data
-    embedding = await asyncio.to_thread(_sync_embed, _embed_text(transcript, incident_type))
+    processed: list[dict] = []
+    for scene in raw_scenes:
+        tags:               list[str]      = scene.get("tags") or []
+        incident_type:      Optional[str]  = scene.get("incident_type") or None
+        location:           Optional[str]  = scene.get("location") or None
+        vehicles:           list[str]      = scene.get("vehicles") or []
+        units:              list[str]      = scene.get("units") or []
+        severity:           str            = scene.get("severity") or "unknown"
+        resolved:           bool           = bool(scene.get("resolved", False))
+        transcript_corrected: Optional[str]= scene.get("transcript_corrected") or None
+        segment_indices:    Optional[list] = scene.get("segment_indices")
 
-    updates: dict = {"tags": tags, "severity": severity}
-    if location:
-        updates["location"] = location
-    if location_coords:
-        updates["location_coords"] = location_coords
-    if vehicles:
-        updates["vehicles"] = vehicles
-    if units:
-        updates["units"] = units
-    if embedding:
-        updates["embedding"] = embedding
-    if transcript_corrected and not preserve_transcript_correction:
-        updates["transcript_corrected"] = transcript_corrected
+        if incident_type in ("unknown", "other", ""):
+            incident_type = None
+
+        # Geocode this scene's location
+        location_coords: Optional[dict] = None
+        if location and node_lat is not None and node_lon is not None:
+            state = await _get_node_state(node_id, node_lat, node_lon)
+            muni  = _municipality_from_tg(talkgroup_name)
+            hint_parts = [p for p in [muni, state] if p]
+            query = f"{location}, {', '.join(hint_parts)}" if hint_parts else location
+            location_coords = await _geocode_location(query, node_lat, node_lon)
+
+        # Embed this scene's content
+        scene_text = _build_scene_embed_text(
+            transcript, segments, segment_indices, incident_type, transcript_corrected
+        )
+        embedding = await asyncio.to_thread(_sync_embed, scene_text)
+
+        processed.append({
+            "tags":                 tags,
+            "incident_type":        incident_type,
+            "location":             location,
+            "location_coords":      location_coords,
+            "vehicles":             vehicles,
+            "units":                units,
+            "severity":             severity,
+            "resolved":             resolved,
+            "transcript_corrected": transcript_corrected,
+            "segment_indices":      segment_indices,
+            "embedding":            embedding,
+        })
+
+    # Merge across scenes for the call-level Firestore document.
+    # Primary scene (first) owns location, severity, transcript_corrected.
+    # Tags/units/vehicles are union-merged from all scenes.
+    primary = processed[0]
+    all_tags     = list(dict.fromkeys(t for s in processed for t in s["tags"]))
+    all_units    = list(dict.fromkeys(u for s in processed for u in s["units"]))
+    all_vehicles = list(dict.fromkeys(v for s in processed for v in s["vehicles"]))
+
+    updates: dict = {"tags": all_tags, "severity": primary["severity"]}
+    if primary["location"]:
+        updates["location"] = primary["location"]
+    if primary["location_coords"]:
+        updates["location_coords"] = primary["location_coords"]
+    if all_units:
+        updates["units"] = all_units
+    if all_vehicles:
+        updates["vehicles"] = all_vehicles
+    if primary["embedding"]:
+        updates["embedding"] = primary["embedding"]
+    if primary["transcript_corrected"] and not preserve_transcript_correction:
+        updates["transcript_corrected"] = primary["transcript_corrected"]
 
     try:
         await fstore.doc_set("calls", call_id, updates)
     except Exception as e:
         logger.warning(f"Could not save intelligence for call {call_id}: {e}")
 
-    logger.info(
-        f"Intelligence: call {call_id} → type={incident_type}, "
-        f"tags={tags}, location={location!r}, coords={location_coords}, severity={severity}, "
-        f"corrected={transcript_corrected is not None}"
+    scene_summary = (
+        f"{len(processed)} scene(s): "
+        + ", ".join(
+            f"[{s['incident_type'] or 'unclassified'} tags={s['tags'][:2]}]"
+            for s in processed
+        )
     )
-    return tags, incident_type, location, location_coords, resolved
+    logger.info(f"Intelligence: call {call_id} → {scene_summary}")
+    return processed
 
 
 async def _geocode_location(
@@ -220,7 +280,6 @@ def _municipality_from_tg(tg_name: Optional[str]) -> Optional[str]:
     if not tg_name:
         return None
     cleaned = _TG_SUFFIX_RE.sub("", tg_name).strip()
-    # Discard if nothing left, purely numeric, or a short all-caps abbreviation (e.g. "WC", "TAC")
     if not cleaned or cleaned.isdigit() or (len(cleaned) <= 3 and cleaned.isupper()):
         return None
     return cleaned
@@ -234,6 +293,23 @@ def _build_transcript_block(transcript: str, segments: Optional[list[dict]]) ->
     return f"Transcript:\n{transcript}"
 
 
+def _build_scene_embed_text(
+    transcript: str,
+    segments: Optional[list[dict]],
+    segment_indices: Optional[list[int]],
+    incident_type: Optional[str],
+    transcript_corrected: Optional[str],
+) -> str:
+    """Build the text string to embed for a specific scene."""
+    prefix = f"[{incident_type}] " if incident_type else ""
+    if transcript_corrected:
+        return f"{prefix}{transcript_corrected}"
+    if segments and segment_indices:
+        texts = [segments[i]["text"] for i in segment_indices if i < len(segments)]
+        return f"{prefix}{' '.join(texts)}"
+    return f"{prefix}{transcript}"
+
+
 def _sync_extract(
     transcript: str,
     talkgroup_name: Optional[str],
@@ -241,14 +317,15 @@ def _sync_extract(
     system_id: Optional[str],
     segments: Optional[list[dict]],
     vocabulary: Optional[list[str]] = None,
-) -> dict:
-    """Call GPT-4o mini and parse the JSON response."""
+    ten_codes: Optional[dict[str, str]] = None,
+) -> list[dict]:
+    """Call GPT-4o-mini and return a list of scene dicts."""
     from app.config import settings
     from openai import OpenAI
 
     if not settings.openai_api_key:
         logger.warning("OPENAI_API_KEY not set — intelligence extraction disabled.")
-        return {}
+        return []
 
     from app.internal.vocabulary_learner import build_gpt_vocab_block
     tg = f"{talkgroup_name} (TGID {talkgroup_id})" if talkgroup_id else (talkgroup_name or "unknown")
@@ -256,6 +333,7 @@ def _sync_extract(
         transcript_block=_build_transcript_block(transcript, segments),
         talkgroup_name=tg,
         system_id=system_id or "unknown",
+        ten_codes_block=_build_ten_codes_block(ten_codes or {}),
         vocabulary_block=build_gpt_vocab_block(vocabulary or []),
     )
 
@@ -266,13 +344,22 @@ def _sync_extract(
             messages=[{"role": "user", "content": prompt}],
             response_format={"type": "json_object"},
         )
-        return json.loads(response.choices[0].message.content)
+        raw = json.loads(response.choices[0].message.content)
+
+        # New format: {"scenes": [...]}
+        if "scenes" in raw and isinstance(raw["scenes"], list):
+            return raw["scenes"]
+
+        # Fallback: GPT returned the old flat single-scene format
+        logger.warning("GPT returned flat format instead of scenes array — wrapping")
+        return [raw]
+
     except json.JSONDecodeError as e:
-        logger.warning(f"GPT-4o mini returned non-JSON: {e}")
-        return {}
+        logger.warning(f"GPT-4o-mini returned non-JSON: {e}")
+        return []
     except Exception as e:
-        logger.warning(f"GPT-4o mini extraction failed: {e}")
-        return {}
+        logger.warning(f"GPT-4o-mini extraction failed: {e}")
+        return []
 
 
 def _sync_embed(text: str) -> Optional[list[float]]:
@@ -290,8 +377,3 @@ def _sync_embed(text: str) -> Optional[list[float]]:
     except Exception as e:
         logger.warning(f"Embedding generation failed: {e}")
         return None
-
-
-def _embed_text(transcript: str, incident_type: Optional[str]) -> str:
-    prefix = f"[{incident_type}] " if incident_type else ""
-    return f"{prefix}{transcript}"

drb-c2-core/app/internal/recorrelation_sweep.py

@@ -46,7 +46,10 @@ async def _run_sweep_pass() -> None:
         ("status",   "==",  "ended"),
         ("ended_at", ">=",  cutoff),
     ])
-    orphans = [c for c in recent_ended if not c.get("incident_id")]
+    orphans = [
+        c for c in recent_ended
+        if not c.get("incident_ids") and not c.get("incident_id")
+    ]
 
     if not orphans:
         return
@@ -89,6 +92,7 @@ async def _recorrelate_orphan(call: dict) -> bool:
     )
 
     if incident_id:
+        await fstore.doc_set("calls", call_id, {"incident_ids": [incident_id]})
         logger.info(
             f"Re-correlation: linked orphaned call {call_id} → incident {incident_id}"
         )

drb-c2-core/app/models.py

@@ -33,12 +33,14 @@ class SystemRecord(BaseModel):
     name: str
     type: str                        # P25 / DMR / NBFM
     config: Dict[str, Any] = {}     # OP25-compatible config blob
+    ten_codes: Dict[str, str] = {}  # {"10-10": "Commercial Alarm", ...}
 
 
 class SystemCreate(BaseModel):
     name: str
     type: str
     config: Dict[str, Any] = {}
+    ten_codes: Dict[str, str] = {}
 
 
 # ---------------------------------------------------------------------------
@@ -56,11 +58,11 @@ class CallRecord(BaseModel):
     started_at: datetime
     ended_at: Optional[datetime] = None
     audio_url: Optional[str] = None
-    transcript: Optional[str] = None   # populated later by STT
-    incident_id: Optional[str] = None  # populated later by intelligence layer
+    transcript: Optional[str] = None      # populated later by STT
+    incident_ids: List[str] = []          # one per scene detected in the recording
     location: Optional[Dict[str, float]] = None  # {lat, lng}
     tags: List[str] = []
-    status: str = "active"             # active / ended
+    status: str = "active"                # active / ended
 
 
 # ---------------------------------------------------------------------------

drb-c2-core/app/routers/calls.py

@@ -83,6 +83,28 @@ async def patch_transcript(
         "embedding": None,
     })
 
+    # Unlink from ALL current incidents so re-correlation starts clean.
+    # Handles both old single incident_id and new incident_ids list.
+    old_ids: list[str] = call.get("incident_ids") or (
+        [call["incident_id"]] if call.get("incident_id") else []
+    )
+    for old_incident_id in old_ids:
+        old_incident = await fstore.doc_get("incidents", old_incident_id)
+        if old_incident:
+            remaining = [c for c in (old_incident.get("call_ids") or []) if c != call_id]
+            if remaining:
+                await fstore.doc_set("incidents", old_incident_id, {
+                    "call_ids": remaining,
+                    "summary_stale": True,
+                })
+            else:
+                await fstore.doc_set("incidents", old_incident_id, {
+                    "call_ids": [],
+                    "status": "resolved",
+                    "summary_stale": True,
+                })
+    await fstore.doc_set("calls", call_id, {"incident_ids": [], "incident_id": None})
+
     # Learn from the correction: diff original → corrected and add new tokens to vocabulary
     system_id = call.get("system_id")
     original_text = call.get("transcript_corrected") or call.get("transcript") or ""

drb-c2-core/app/routers/systems.py

@@ -1,7 +1,7 @@
 import uuid
 from fastapi import APIRouter, HTTPException
 from pydantic import BaseModel
-from typing import Optional
+from typing import Dict, Optional
 from app.models import SystemCreate, SystemRecord
 from app.internal import firestore as fstore
 
@@ -12,6 +12,10 @@ class VocabularyTermBody(BaseModel):
     term: str
 
 
+class TenCodesBody(BaseModel):
+    ten_codes: Dict[str, str]
+
+
 @router.get("")
 async def list_systems():
     return await fstore.collection_list("systems")
@@ -50,6 +54,27 @@ async def delete_system(system_id: str):
     await fstore.doc_delete("systems", system_id)
 
 
+# ── Ten-codes endpoints ────────────────────────────────────────────────────────
+
+@router.get("/{system_id}/ten-codes")
+async def get_ten_codes(system_id: str):
+    """Return the ten-code dictionary for a system."""
+    system = await fstore.doc_get("systems", system_id)
+    if not system:
+        raise HTTPException(404, f"System '{system_id}' not found.")
+    return {"ten_codes": system.get("ten_codes") or {}}
+
+
+@router.put("/{system_id}/ten-codes")
+async def update_ten_codes(system_id: str, body: TenCodesBody):
+    """Replace the ten-code dictionary for a system."""
+    existing = await fstore.doc_get("systems", system_id)
+    if not existing:
+        raise HTTPException(404, f"System '{system_id}' not found.")
+    await fstore.doc_update("systems", system_id, {"ten_codes": body.ten_codes})
+    return {"ok": True, "ten_codes": body.ten_codes}
+
+
 # ── Vocabulary endpoints ───────────────────────────────────────────────────────
 
 @router.get("/{system_id}/vocabulary")

drb-c2-core/app/routers/upload.py

@@ -96,35 +96,47 @@ async def _run_extraction_pipeline(
     """Run steps 2-4 of the intelligence pipeline using an existing transcript."""
     from app.internal import intelligence, incident_correlator, alerter
 
-    tags, incident_type, location, location_coords, resolved = await intelligence.extract_tags(
+    # Step 2: Scene detection + intelligence extraction.
+    # Returns one scene per distinct incident detected in the recording.
+    scenes = await intelligence.extract_scenes(
         call_id, transcript, talkgroup_name,
         talkgroup_id=talkgroup_id, system_id=system_id, segments=segments,
         node_id=node_id,
         preserve_transcript_correction=preserve_transcript_correction,
     )
 
-    incident_id = await incident_correlator.correlate_call(
-        call_id=call_id,
-        node_id=node_id,
-        system_id=system_id,
-        talkgroup_id=talkgroup_id,
-        talkgroup_name=talkgroup_name,
-        tags=tags,
-        incident_type=incident_type,
-        location=location,
-        location_coords=location_coords,
-    )
+    # Step 3: Correlate each scene to an incident independently.
+    incident_ids: list[str] = []
+    all_tags: list[str] = []
+    for scene in scenes:
+        all_tags.extend(scene["tags"])
+        incident_id = await incident_correlator.correlate_call(
+            call_id=call_id,
+            node_id=node_id,
+            system_id=system_id,
+            talkgroup_id=talkgroup_id,
+            talkgroup_name=talkgroup_name,
+            tags=scene["tags"],
+            incident_type=scene["incident_type"],
+            location=scene["location"],
+            location_coords=scene["location_coords"],
+        )
+        if incident_id and incident_id not in incident_ids:
+            incident_ids.append(incident_id)
+        if scene["resolved"] and incident_id:
+            await fstore.doc_set("incidents", incident_id, {"status": "resolved"})
+            logger.info(f"Auto-resolved incident {incident_id} (LLM closure detection)")
 
-    if resolved and incident_id:
-        await fstore.doc_set("incidents", incident_id, {"status": "resolved"})
-        logger.info(f"Auto-resolved incident {incident_id} (LLM closure detection)")
+    if incident_ids:
+        await fstore.doc_set("calls", call_id, {"incident_ids": incident_ids})
 
+    # Step 4: Alert dispatch — run once with merged tags from all scenes.
     await alerter.check_and_dispatch(
         call_id=call_id,
         node_id=node_id,
         talkgroup_id=talkgroup_id,
         talkgroup_name=talkgroup_name,
-        tags=tags,
+        tags=list(dict.fromkeys(all_tags)),
         transcript=transcript,
     )
 
@@ -140,8 +152,8 @@ async def _run_intelligence_pipeline(
     """
     Post-upload intelligence pipeline (runs as a background task):
       1. Transcribe audio via Google STT
-      2. Extract tags/incident type from transcript
-      3. Correlate with existing incidents (or create new one)
+      2. Detect scenes + extract intelligence (one result per incident in recording)
+      3. Correlate each scene with existing incidents (or create new ones)
       4. Check alert rules and dispatch notifications
     """
     from app.internal import transcription, intelligence, incident_correlator, alerter
@@ -155,35 +167,57 @@ async def _run_intelligence_pipeline(
             call_id, gcs_uri, talkgroup_name, system_id=system_id
         )
 
-    # Step 2: Intelligence extraction
-    tags: list[str] = []
-    incident_type: Optional[str] = None
-    location: Optional[str] = None
-    location_coords: Optional[dict] = None
-    resolved: bool = False
+    # Step 2: Scene detection + intelligence extraction
+    scenes: list[dict] = []
     if transcript:
-        tags, incident_type, location, location_coords, resolved = await intelligence.extract_tags(
+        scenes = await intelligence.extract_scenes(
             call_id, transcript, talkgroup_name,
             talkgroup_id=talkgroup_id, system_id=system_id, segments=segments,
             node_id=node_id,
         )
 
-    # Step 3: Incident correlation (always runs — unclassified calls can still link via talkgroup)
-    incident_id = await incident_correlator.correlate_call(
-        call_id=call_id,
-        node_id=node_id,
-        system_id=system_id,
-        talkgroup_id=talkgroup_id,
-        talkgroup_name=talkgroup_name,
-        tags=tags,
-        incident_type=incident_type,
-        location=location,
-        location_coords=location_coords,
-    )
+    # Step 3: Correlate each scene independently.
+    # A single recording can produce multiple incidents on a busy channel.
+    incident_ids: list[str] = []
+    all_tags: list[str] = []
+    for scene in scenes:
+        all_tags.extend(scene["tags"])
+        incident_id = await incident_correlator.correlate_call(
+            call_id=call_id,
+            node_id=node_id,
+            system_id=system_id,
+            talkgroup_id=talkgroup_id,
+            talkgroup_name=talkgroup_name,
+            tags=scene["tags"],
+            incident_type=scene["incident_type"],
+            location=scene["location"],
+            location_coords=scene["location_coords"],
+        )
+        if incident_id and incident_id not in incident_ids:
+            incident_ids.append(incident_id)
+        if scene["resolved"] and incident_id:
+            await fstore.doc_set("incidents", incident_id, {"status": "resolved"})
+            logger.info(f"Auto-resolved incident {incident_id} (LLM closure detection)")
 
-    if resolved and incident_id:
-        await fstore.doc_set("incidents", incident_id, {"status": "resolved"})
-        logger.info(f"Auto-resolved incident {incident_id} (LLM closure detection)")
+    # Correlator also runs for calls with no scenes (unclassified) to attempt
+    # talkgroup-based linking even when no transcript could be produced.
+    if not scenes:
+        incident_id = await incident_correlator.correlate_call(
+            call_id=call_id,
+            node_id=node_id,
+            system_id=system_id,
+            talkgroup_id=talkgroup_id,
+            talkgroup_name=talkgroup_name,
+            tags=[],
+            incident_type=None,
+            location=None,
+            location_coords=None,
+        )
+        if incident_id:
+            incident_ids.append(incident_id)
+
+    if incident_ids:
+        await fstore.doc_set("calls", call_id, {"incident_ids": incident_ids})
 
     # Step 4: Alert dispatch (always runs — talkgroup ID rules don't need a transcript)
     await alerter.check_and_dispatch(
@@ -191,6 +225,6 @@ async def _run_intelligence_pipeline(
         node_id=node_id,
         talkgroup_id=talkgroup_id,
         talkgroup_name=talkgroup_name,
-        tags=tags,
+        tags=list(dict.fromkeys(all_tags)),
         transcript=transcript,
     )