redeem fix

.github/workflows/build-and-release.yml

@@ -617,6 +617,7 @@ jobs:
   docker-build:
     name: Build and Push Docker Image
     runs-on: ubuntu-latest
+    if: false  # Disabled - Docker build step is commented out; re-enable when build-push-action is restored
     continue-on-error: true
     needs: build  # Run after Windows build succeeds
 

CHANGELOG.md

@@ -0,0 +1,67 @@
+# Changelog
+
+All notable changes to Chat Yapper are documented here.
+
+## v1.3.3 (Latest)
+- twitch redeem fix
+- two new voice effects
+- new filter for @user messages
+
+## v1.3.2
+- Fix to edge-tts
+
+## v1.3.1
+- **New Features:**
+  - Twitch fix
+  - Allow random avatar assignment
+  - More fonts to select from
+
+## v1.3.0
+- Better control of avatar placement in Avatar Layout Editor
+- Select and Adjust speaking animations for crowd mode
+- Added idle animations for crowd mode
+
+## v1.2.2
+- Quick status view
+- Limit concurrent TTS messages
+- Some more twitch fixes and improved notifications
+
+## v1.2.1
+- Usernames for chat bubbles
+- Text size adjustment
+- Toggle for only allowing redeem messages for Twitch
+- Twitch token refresh (should fix auth issues)
+
+## v1.2.0
+- Chat bubbles above avatars
+- Pop-up mode for avatars
+- Linux standalone build (x64)
+- Improved audio quality (reduced crackling)
+- Docker multi-architecture support (amd64, arm64)
+- Better audio preloading and buffering
+- High-quality ffmpeg audio processing
+- GitHub Container Registry (GHCR) for Docker images
+- Automated cross-platform builds via GitHub Actions
+
+## v1.1.2
+- Stability fixes
+- MSI installation for Windows
+- Light mode theme
+- Cleaned up settings UI
+
+## v1.0.0
+- GIF and WebP support for animated avatars
+- Customizable speaking glow effects (color, opacity, size, enable/disable)
+- Message history and replay system (stores 100 recent messages)
+- Export/Import configuration system (backup/restore settings, voices, and avatars)
+- Persistent voice caching for all TTS providers (MonsterTTS, Google Cloud, Amazon Polly, Edge TTS)
+- Audio filters
+- Docker Support
+
+## v0.1.0
+- Initial release
+- Avatars page
+- Settings page
+- TTS selection
+- Avatar positioning
+- Basic Twitch integration

README.md

@@ -225,64 +225,7 @@ chat-yapper/
 
 ## Changelog
 
-### v1.3.2 (Latest)
- - Fix to edge-tts
-
-### v1.3.1
-- **New Features:**
- - Twitch fix
- - Allow random avatar assignment
- - More fonts to select from
-
-### v1.3.0
- - Better control of avatar placement in Avatar Layout Editor
- - Select and Adjust speaking animations for crowd mode
- - Added idle animations for crowd mode
-
-### v1.2.2
- - Quick status view
- - Limit concurrent TTS messages
- - Some more twitch fixes and improved notifcations
-
-### v1.2.1
- - usernames for chatbubbles
- - text size adjustment
- - Toggle for only allowing redeem messages for twitch
- - twitch token refresh (should fix auth issues)
-
-### v1.2.0
-- Chat bubbles above avatars
-- Pop-up mode for avatars
-- Linux standalone build (x64)
-- Improved audio quality (reduced crackling)
-- Docker multi-architecture support (amd64, arm64) 
-- Better audio preloading and buffering
-- High-quality ffmpeg audio processing
-- GitHub Container Registry (GHCR) for Docker images
-- Automated cross-platform builds via GitHub Actions
-
-### v1.1.2
-- Stability fixes
-- MSI installation for Windows
-- Light mode theme
-- Cleaned up settings UI
-
-### v1.0.0
-- GIF and WebP support for animated avatars
-- Customizable speaking glow effects (color, opacity, size, enable/disable)
-- Message history and replay system (stores 100 recent messages)
-- Export/Import configuration system (backup/restore settings, voices, and avatars)
-- Persistent voice caching for all TTS providers (MonsterTTS, Google Cloud, Amazon Polly, Edge TTS)
-- Audio filters
-- Docker Support
-
-### v0.1.0
-- Initial release
-- Avatars page
-- Settings page
-- TTS selection
-- Avatar positioning
-- Basic Twitch integration
+See [CHANGELOG.md](CHANGELOG.md) for the full version history.
 
 ---
 
@@ -319,6 +262,7 @@ This application was inspired by the work done by [shindigs](https://x.com/shind
 Special thanks these streamers that helped test the early prototype:
 - [**Inislein**](https://x.com/iniskein)
 - [**Kirana**](https://x.com/KiranaYonome)
+- [**Miaelou**](https://x.com/Miaelou_VT)
 
 ---
 

backend/modules/audio_filters.py

@@ -178,6 +178,8 @@ def _has_enabled_filters(self, settings: Dict[str, Any]) -> bool:
             settings.get("reverb", {}).get("enabled", False),
             settings.get("pitch", {}).get("enabled", False),
             settings.get("speed", {}).get("enabled", False),
+            settings.get("underwater", {}).get("enabled", False),
+            settings.get("vibrato", {}).get("enabled", False),
         ])
 
     def _build_filters(self, settings: Dict[str, Any]) -> List[str]:
@@ -212,7 +214,27 @@ def _build_filters(self, settings: Dict[str, Any]) -> List[str]:
                 else:  # speed > 2.0
                     # Chain multiple atempo for very fast speeds
                     filters.append(f"atempo=2.0,atempo={speed/2.0}")
-
+
+        # Underwater effect
+        # Technique: low-pass filter (water absorbs high freqs) + echo (watery reverb) + chorus (wobbly sub-surface)
+        if settings.get("underwater", {}).get("enabled", False):
+            intensity = settings.get("underwater", {}).get("intensity", 50)  # 0-100
+            # Map intensity to low-pass cutoff: 0% = 1200 Hz (shallow), 100% = 500 Hz (deep)
+            freq = int(1200 - (intensity / 100.0) * 700)
+            filters.append(
+                f"lowpass=f={freq},"
+                f"aecho=0.8:0.88:60|80:0.5|0.4,"
+                f"chorus=0.7:0.9:55|60:0.4|0.35:0.25|0.4:2|1.3"
+            )
+
+        # Vibrato effect: periodic pitch modulation
+        if settings.get("vibrato", {}).get("enabled", False):
+            rate = settings.get("vibrato", {}).get("rate", 10.0)  # Hz, 6.0-15.0
+            depth = settings.get("vibrato", {}).get("depth", 75) / 100.0  # 0.0-1.0
+            rate = max(0.1, min(20.0, float(rate)))
+            depth = max(0.0, min(1.0, depth))
+            filters.append(f"vibrato=f={rate:.1f}:d={depth:.2f}")
+
         return filters
 
     def _build_random_filters(self, settings: Dict[str, Any]) -> List[str]:
@@ -228,6 +250,11 @@ def _build_random_filters(self, settings: Dict[str, Any]) -> List[str]:
         if settings.get("speed", {}).get("randomEnabled", True):
             available_filters.append("speed")
 
+        if settings.get("underwater", {}).get("randomEnabled", True):
+            available_filters.append("underwater")
+        if settings.get("vibrato", {}).get("randomEnabled", True):
+            available_filters.append("vibrato")
+
         if not available_filters:
             logger.warning("No effects enabled for random mode")
             return filters
@@ -239,7 +266,23 @@ def _build_random_filters(self, settings: Dict[str, Any]) -> List[str]:
         selected = random.sample(available_filters, num_filters)
 
         for filter_type in selected:
-            if filter_type == "reverb":
+            if filter_type == "underwater":
+                intensity = random.randint(30, 80)
+                freq = int(1200 - (intensity / 100.0) * 700)
+                filters.append(
+                    f"lowpass=f={freq},"
+                    f"aecho=0.8:0.88:60|80:0.5|0.4,"
+                    f"chorus=0.7:0.9:55|60:0.4|0.35:0.25|0.4:2|1.3"
+                )
+                logger.debug(f"Random underwater: intensity={intensity}% (cutoff={freq} Hz)")
+
+            elif filter_type == "vibrato":
+                rate = round(random.uniform(6.0, 15.0), 1)
+                depth = round(random.uniform(0.5, 1.0), 2)
+                filters.append(f"vibrato=f={rate}:d={depth}")
+                logger.debug(f"Random vibrato: rate={rate} Hz, depth={depth}")
+
+            elif filter_type == "reverb":
                 # Get custom range or use defaults
                 reverb_config = settings.get("reverb", {})
                 random_range = reverb_config.get("randomRange", {"min": 20, "max": 80})

backend/modules/message_filter.py

@@ -138,33 +138,31 @@ def should_process_message(
     Returns:
         (should_process, filtered_text) - tuple indicating if message should be processed and the filtered text
     """
-    filtering = settings.get("messageFiltering", {})
-
-    if not filtering.get("enabled", True):
-        return True, text
-
-    # Check Twitch channel point redeem filter
+    # Check Twitch channel point redeem filter first — this applies regardless of
+    # whether general message filtering is enabled or disabled.
     twitch_settings = settings.get("twitch", {})
     redeem_filter = twitch_settings.get("redeemFilter", {})
     if redeem_filter.get("enabled", False):
-        allowed_redeem_names = redeem_filter.get("allowedRedeemNames", [])
-        if allowed_redeem_names:
-            # Check if message has a msg-param-reward-name tag (the redeem name)
-            # Also check custom-reward-id to confirm it's a redeem
-            custom_reward_id = tags.get("custom-reward-id", "") if tags else ""
-            reward_name = tags.get("msg-param-reward-name", "") if tags else ""
-
-            if not custom_reward_id:
-                # No redeem ID means this is a regular message, not a channel point redeem
-                logger.info(f"Skipping message from {username} - not from a channel point redeem")
-                return False, text
-
-            # Check if the redeem name is in the allowed list (case-insensitive)
-            if not any(reward_name.lower() == allowed_name.lower() for allowed_name in allowed_redeem_names):
-                logger.info(f"Skipping message from {username} - redeem name '{reward_name}' not in allowed list")
-                return False, text
-
-            logger.info(f"Processing message from {username} - redeem name '{reward_name}' is allowed")
+        # Twitch IRC PRIVMSG tags include custom-reward-id (UUID) for channel point redeems,
+        # but do NOT include the reward title/name. Name-based filtering is not possible
+        # from IRC events alone, so we filter only on the presence of custom-reward-id.
+        # Exception: the built-in "Highlight My Message" reward uses msg-id=highlighted-message
+        # instead of custom-reward-id, so we treat that as a valid redeem too.
+        custom_reward_id = tags.get("custom-reward-id", "") if tags else ""
+        msg_id = (tags.get("msg-id", "") or "") if tags else ""
+        is_highlight = msg_id.lower() == "highlighted-message"
+
+        if not custom_reward_id and not is_highlight:
+            # No redeem ID means this is a regular chat message, not a channel point redeem
+            logger.info(f"Skipping message from {username} - not from a channel point redeem")
+            return False, text
+
+        logger.info(f"Processing channel point redeem from {username} (reward-id: {custom_reward_id or 'highlighted-message'})")
+
+    filtering = settings.get("messageFiltering", {})
+
+    if not filtering.get("enabled", True):
+        return True, text
 
     # Skip ignored users
     if username and filtering.get("ignoredUsers"):
@@ -180,6 +178,12 @@ def should_process_message(
         if stripped.startswith('!') or stripped.startswith('/'):
             logger.info(f"Skipping command message: {text[:50]}...")
             return False, text
+
+    # Skip messages that @mention someone
+    if filtering.get("skipMentions", False):
+        if re.search(r'@\w+', text):
+            logger.info(f"Skipping mention message from {username}: {text[:50]}...")
+            return False, text
 
     # Start with original text, apply filters progressively
     filtered_text = text

backend/modules/settings_defaults.json

@@ -73,6 +73,7 @@
 "minLength": 1,
 "maxLength": 500,
 "skipCommands": true,
+"skipMentions": false,
 "skipEmotes": false,
 "removeUrls": true,
 "ignoredUsers": [],
@@ -118,6 +119,17 @@
 "min": 0.75,
 "max": 1.3
 }
+},
+"underwater": {
+"enabled": false,
+"intensity": 50,
+"randomEnabled": true
+},
+"vibrato": {
+"enabled": false,
+"rate": 10.0,
+"depth": 75,
+"randomEnabled": true
 }
 }
 }

backend/modules/tts.py

@@ -22,54 +22,25 @@ def reset_fallback_stats():
     fallback_voice_stats.clear()
     fallback_selection_count = 0
 
-# Track edge-tts update attempts to avoid repeated updates
+# Track edge-tts update attempts to avoid repeated update instructions
 _edge_tts_update_attempted = False
+_edge_tts_update_lock = asyncio.Lock()
 
 async def try_update_edge_tts():
-    """Attempt to update edge-tts package when API compatibility issues occur"""
+    """Log a one-time instruction to update edge-tts outside the running process."""
     global _edge_tts_update_attempted
-
-    if _edge_tts_update_attempted:
-        logger.info("edge-tts update already attempted this session, skipping")
-        return False
-
-    _edge_tts_update_attempted = True
-    logger.info("Attempting to update edge-tts package to fix API compatibility...")
-
-    try:
-        # Run pip upgrade in subprocess
-        python_exe = sys.executable
-        result = await asyncio.create_subprocess_exec(
-            python_exe, "-m", "pip", "install", "--upgrade", "edge-tts",
-            stdout=asyncio.subprocess.PIPE,
-            stderr=asyncio.subprocess.PIPE
-        )
-        stdout, stderr = await result.communicate()
-
-        if result.returncode == 0:
-            logger.info(f"edge-tts successfully updated: {stdout.decode()}")
-
-            # Reload the edge_tts module
-            try:
-                import importlib
-                global edge_tts
-                if edge_tts:
-                    importlib.reload(edge_tts)
-                    logger.info("edge-tts module reloaded successfully")
-                else:
-                    import edge_tts as new_edge_tts
-                    edge_tts = new_edge_tts
-                    logger.info("edge-tts module imported successfully")
-                return True
-            except Exception as e:
-                logger.warning(f"edge-tts updated but module reload failed: {e}")
-                logger.info("Restart the application to use the updated edge-tts")
-                return False
-        else:
-            logger.error(f"edge-tts update failed: {stderr.decode()}")
+
+    async with _edge_tts_update_lock:
+        if _edge_tts_update_attempted:
+            logger.info("edge-tts update already flagged this session, skipping")
             return False
-    except Exception as e:
-        logger.error(f"Failed to update edge-tts: {e}")
+
+        _edge_tts_update_attempted = True
+        logger.error(
+            "edge-tts appears incompatible, but automatic runtime upgrades are disabled. "
+            "Please update the pinned 'edge-tts' dependency through your normal build/deploy "
+            "process and restart the application."
+        )
         return False
 
 # Provider 1: MonsterAPI TTS (async, great quality)