feat: implement accumulative LLM paragraph engine and harden 4-process pipeline

engine_distribute.py

@@ -3,11 +3,22 @@ import time
 import argparse
 import multiprocessing
 import sys
+import os
 
 INGEST_URL = "https://emiapi.reynafamily.com/live-captions/ingest"
 
+def log_debug(message):
+    try:
+        with open("distribute_debug.log", "a") as f:
+            f.write(f"{time.strftime('%H:%M:%S')} {message}\n")
+            f.flush()
+            os.fsync(f.fileno())
+    except:
+        pass
+
 def run_distribution(in_queue, args):
-    print("[Distribute] Starting distribution service...")
+    print("[Distribute] Starting distribution service...", flush=True)
+    log_debug("Starting distribution service...")
     
     while True:
         try:
@@ -20,14 +31,30 @@ def run_distribution(in_queue, args):
                 sys.stdout.write(f"\r\033[K[DRAFT]: {draft_text}")
                 sys.stdout.flush()
                 if args.ingest:
-                    requests.post(INGEST_URL, json={"draft": draft_text}, timeout=2)
+                    try: requests.post(INGEST_URL, json={"draft": draft_text}, timeout=2)
+                    except: pass
                 continue
 
             # Finalized segment
-            print(f"\n[Final]: {payload.get('original', '')}")
+            speaker_label = payload.get('speaker')
+            speaker_prefix = f"[{speaker_label}]: " if speaker_label else ""
+            
+            msg = f"\n[Final]: {speaker_prefix}{payload.get('original', '')}"
+            print(msg, flush=True)
+            log_debug(msg)
+
+            # Show LLM Refinement if available
+            refined = payload.get('paragraph') or payload.get('corrected')
+            if refined and refined.strip() != payload.get('original', '').strip():
+                llm_msg = f"[LLM]: {speaker_prefix}{refined}"
+                print(llm_msg, flush=True)
+                log_debug(llm_msg)
+            
             for lang in ["es", "fr", "ar", "en"]:
                 if payload.get(lang):
-                    print(f"[{lang.upper()}]: {payload[lang]}")
+                    lang_msg = f"[{lang.upper()}]: {speaker_prefix}{payload[lang]}"
+                    print(lang_msg, flush=True)
+                    log_debug(lang_msg)
 
             if args.ingest:
                 delay, max_delay, success = 1, 15, False