stt updates and intelligence updates

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

@@ -11,17 +11,34 @@ from typing import Optional
 from app.internal.logger import logger
 from app.internal import firestore as fstore
 
+# Whisper treats `prompt` as preceding transcript text, not instructions.
+# Writing it as actual radio speech primes the vocabulary toward P25 codes
+# and phrasing before the model hears the audio.
+_WHISPER_PROMPT = (
+    "10-4. 10-23. 10-20. 10-97. 10-8. 10-7. 10-34. 10-50. 10-52. "
+    "Post 4, I'm out. Post 3. En route. On scene. In route. "
+    "Copy. Negative. Stand by. Be advised. Go ahead. "
+    "Units responding. Dispatch. Talkgroup. "
+    "Engine. Ladder. Medic. Rescue. Car. Unit. "
+    "MVA. MVC. Structure fire. Working fire."
+)
 
-async def transcribe_call(call_id: str, gcs_uri: str) -> Optional[str]:
+
+async def transcribe_call(
+    call_id: str,
+    gcs_uri: str,
+    talkgroup_name: Optional[str] = None,
+) -> Optional[str]:
     """
     Transcribe audio at the given GCS URI and store the result in Firestore.
 
     Args:
-        call_id: Firestore document ID in the 'calls' collection.
-        gcs_uri:  GCS URI of the audio file, e.g. gs://bucket/calls/xyz.mp3
+        call_id:        Firestore document ID in the 'calls' collection.
+        gcs_uri:        GCS URI of the audio file, e.g. gs://bucket/calls/xyz.mp3
+        talkgroup_name: Passed through to the intelligence layer; unused here.
 
     Returns:
-        The transcript string, or None if transcription failed / was skipped.
+        The raw Whisper transcript string, or None if transcription failed.
     """
     if not gcs_uri or not gcs_uri.startswith("gs://"):
         return None
@@ -53,11 +70,9 @@ def _sync_transcribe(gcs_uri: str) -> Optional[str]:
         logger.warning("OPENAI_API_KEY not set — transcription disabled.")
         return None
 
-    # Parse gs://bucket/path/to/file.mp3
     without_scheme = gcs_uri[len("gs://"):]
     bucket_name, blob_path = without_scheme.split("/", 1)
 
-    # Download to a temp file
     if settings.gcp_credentials_path:
         creds = service_account.Credentials.from_service_account_file(
             settings.gcp_credentials_path,
@@ -83,7 +98,7 @@ def _sync_transcribe(gcs_uri: str) -> Optional[str]:
                 model="whisper-1",
                 file=f,
                 language="en",
-                prompt="Public safety radio communication. May include police codes, fire, EMS, talkgroup IDs, unit numbers, addresses.",
+                prompt=_WHISPER_PROMPT,
             )
         return response.text.strip() or None
     finally: