fix: Correct TTS endpoint and payload format, filter .mp4 only

modules/text_story/services/tts_handler.py

@@ -1,6 +1,6 @@
 """
 TTS Handler for Text Story module.
-Handles voice generation and audio processing.
+Handles voice generation and audio processing using Kokoro TTS.
 """
 
 import os
@@ -21,6 +21,10 @@ class TTSHandler:
         self.tts_url = os.getenv("HF_TTS", "")
         if not self.tts_url:
             logger.warning("TTSHandler: HF_TTS not configured, TTS will fail")
+        else:
+            # Remove trailing slash
+            self.tts_url = self.tts_url.rstrip('/')
+            logger.info(f"TTSHandler: Using TTS endpoint {self.tts_url}")
     
     async def generate_tts(self, text: str, voice: str, output_path: str) -> float:
         """
@@ -37,23 +41,32 @@ class TTSHandler:
         if not self.tts_url:
             raise ValueError("HF_TTS environment variable not set")
         
+        # Correct endpoint format (same as video_creator)
+        endpoint = f"{self.tts_url}/v1/audio/speech"
+        
+        logger.info(f"TTS: Generating voice '{voice}' for: {text[:50]}...")
+        
         try:
             async with aiohttp.ClientSession() as session:
+                # Correct payload format for Kokoro TTS
                 payload = {
-                    "text": text,
+                    "model": "kokoro",
+                    "input": text,
                     "voice": voice
                 }
                 
                 async with session.post(
-                    f"{self.tts_url}/tts",
+                    endpoint,
                     json=payload,
-                    timeout=aiohttp.ClientTimeout(total=60)
+                    headers={"Content-Type": "application/json"},
+                    timeout=aiohttp.ClientTimeout(total=120)
                 ) as response:
                     if response.status != 200:
                         error_text = await response.text()
-                        raise Exception(f"TTS failed: {error_text}")
+                        raise Exception(f"TTS API error ({response.status}): {error_text}")
                     
                     audio_data = await response.read()
+                    logger.info(f"TTS: Received {len(audio_data)} bytes")
                     
                     # Save raw audio
                     temp_path = output_path + ".temp.wav"
@@ -70,8 +83,11 @@ class TTSHandler:
                     logger.info(f"TTS: Generated {len(text)} chars, {duration:.2f}s")
                     return duration
                     
+        except aiohttp.ClientError as e:
+            logger.error(f"TTS network error: {type(e).__name__}: {e}")
+            raise Exception(f"TTS network error: {e}")
         except Exception as e:
-            logger.error(f"TTS generation failed: {e}")
+            logger.error(f"TTS generation failed: {type(e).__name__}: {e}")
             raise
     
     def trim_silence(self, input_path: str, output_path: str,