diff --git a/.github/workflows/release.yml b/.github/workflows/release.yml
index 04ade01e..1129349f 100644
--- a/.github/workflows/release.yml
+++ b/.github/workflows/release.yml
@@ -309,32 +309,22 @@ jobs:
     steps:
       - name: Checkout repository
         uses: actions/checkout@v6
+        with:
+          fetch-depth: 0  # Full history needed for git-cliff
 
-      - name: Extract changelog for version
+      - name: Generate changelog with git-cliff
         id: changelog
+        uses: orhun/git-cliff-action@v4
+        with:
+          config: cliff.toml
+          args: --latest --strip header
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          OUTPUT: /tmp/changelog.txt
+
+      - name: Show generated changelog
         run: |
-          VERSION=${{ needs.get-version.outputs.version }}
-          VERSION_NUM=${VERSION#v}  # Remove 'v' prefix
-
-          echo "Looking for version: $VERSION_NUM"
-
-          # Extract changelog section for this version using sed
-          # Find the line with version, then print until next version header or end
-          CHANGELOG=$(sed -n "/^## \[$VERSION_NUM\]/,/^## \[/{ /^## \[$VERSION_NUM\]/d; /^## \[/d; p; }" CHANGELOG.md)
-
-          # If no changelog found, use default message
-          if [ -z "$CHANGELOG" ]; then
-            echo "No changelog found for version $VERSION_NUM"
-            CHANGELOG="See CHANGELOG.md for details."
-          else
-            echo "Found changelog content"
-            # Remove trailing --- separator if present (CHANGELOG uses --- between versions)
-            CHANGELOG=$(echo "$CHANGELOG" | sed '/^---$/d')
-          fi
-
-          # Save to file for multiline support
-          echo "$CHANGELOG" > /tmp/changelog.txt
-          echo "Extracted changelog:"
+          echo "Generated changelog:"
           cat /tmp/changelog.txt
 
       - name: Download Android APK
@@ -352,15 +342,13 @@ jobs:
       - name: Prepare release body
         run: |
           VERSION=${{ needs.get-version.outputs.version }}
-          cat > /tmp/release_body.txt << 'HEADER'
-          ### What's New
-          HEADER
-
-          cat /tmp/changelog.txt >> /tmp/release_body.txt
-
           REPO_OWNER="${{ github.repository_owner }}"
           REPO_NAME="${{ github.event.repository.name }}"
-          
+
+          # Start with git-cliff changelog
+          cp /tmp/changelog.txt /tmp/release_body.txt
+
+          # Append download section
           cat >> /tmp/release_body.txt << FOOTER
 
           ---
@@ -404,6 +392,8 @@ jobs:
     steps:
       - name: Checkout repository
         uses: actions/checkout@v6
+        with:
+          fetch-depth: 0
 
       - name: Download Android APK
         uses: actions/download-artifact@v7
@@ -417,52 +407,40 @@ jobs:
           name: ios-ipa
           path: ./release
 
-      - name: Extract changelog for version
+      - name: Generate changelog with git-cliff for Telegram
+        uses: orhun/git-cliff-action@v4
+        with:
+          config: cliff.toml
+          args: --latest --strip all
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          OUTPUT: /tmp/cliff_tg.txt
+
+      - name: Convert changelog for Telegram
         id: changelog
         run: |
-          VERSION=${{ needs.get-version.outputs.version }}
-          VERSION_NUM=${VERSION#v}
-
-          # Extract changelog, limit to ~2500 chars for Telegram (4096 limit minus message overhead)
-          # Use tr -d '\r' to handle CRLF line endings from Windows
-          FULL_CHANGELOG=$(cat CHANGELOG.md | tr -d '\r' | sed -n "/^## \[$VERSION_NUM\]/,/^## \[/{ /^## \[$VERSION_NUM\]/d; /^## \[/d; p; }" | sed '/^---$/d')
-          
-          echo "DEBUG: Extracted changelog length: ${#FULL_CHANGELOG}"
-          echo "DEBUG: First 200 chars: ${FULL_CHANGELOG:0:200}"
-          
-          if [ -z "$FULL_CHANGELOG" ]; then
-            CHANGELOG="See release notes on GitHub for details."
+          if [ ! -s /tmp/cliff_tg.txt ]; then
+            echo "See release notes on GitHub for details." > /tmp/changelog.txt
           else
-            # Convert GitHub Markdown to Telegram HTML:
-            # - **text** → <b>text</b>
-            # - `code` → <code>code</code>
-            # - ### Header → <b>Header</b>
-            # - Escape HTML special chars first
-            # - Remove > blockquote prefix
-            CHANGELOG=$(echo "$FULL_CHANGELOG" | \
-              sed 's/^> //' | \
+            # Convert Markdown to Telegram HTML
+            CHANGELOG=$(cat /tmp/cliff_tg.txt | \
+              sed '/^\*\*Full Changelog\*\*/d' | \
+              sed 's/\[#\([0-9]*\)\]([^)]*)/#\1/g' | \
+              sed 's/\[@\([^]]*\)\]([^)]*)/@\1/g' | \
               sed 's/&/\&amp;/g' | \
               sed 's/</\&lt;/g' | \
               sed 's/>/\&gt;/g' | \
-              sed 's/`\([^`]*\)`/<code>\1<\/code>/g' | \
               sed 's/\*\*\([^*]*\)\*\*/<b>\1<\/b>/g' | \
               sed 's/^### \(.*\)$/<b>\1<\/b>/g' | \
               sed 's/^## \(.*\)$/<b>\1<\/b>/g' | \
-              sed 's/^- /• /g' | \
-              sed 's/^  - /  ◦ /g')
-            
-            # Take first 2500 characters, then cut at last complete line
+              sed 's/^- /• /g')
+
+            # Truncate for Telegram 4096 char limit
             CHANGELOG=$(echo "$CHANGELOG" | head -c 2500 | sed '$d')
-            
-            # Check if truncated
-            FULL_LEN=${#FULL_CHANGELOG}
-            if [ $FULL_LEN -gt 2500 ]; then
-              CHANGELOG="${CHANGELOG}"$'\n\n... (see full changelog on GitHub)'
-            fi
+            echo "$CHANGELOG" > /tmp/changelog.txt
           fi
 
-          echo "$CHANGELOG" > /tmp/changelog.txt
-          echo "DEBUG: Final changelog:"
+          echo "Telegram changelog:"
           cat /tmp/changelog.txt
 
       - name: Send to Telegram Channel
diff --git a/cliff.toml b/cliff.toml
new file mode 100644
index 00000000..0509818b
--- /dev/null
+++ b/cliff.toml
@@ -0,0 +1,105 @@
+# git-cliff configuration for SpotiFLAC Mobile
+# https://git-cliff.org/docs/configuration
+
+[changelog]
+# Template for the changelog body
+body = """
+{%- macro remote_url() -%}
+  https://github.com/zarzet/SpotiFLAC-Mobile
+{%- endmacro -%}
+
+{% if version %}\
+    ## {{ version | trim_start_matches(pat="v") }}
+{% else %}\
+    ## Unreleased
+{% endif %}\
+
+{% for group, commits in commits | group_by(attribute="group") %}
+    ### {{ group | striptags | trim | upper_first }}
+    {% for commit in commits %}
+        - {% if commit.scope %}**{{ commit.scope }}**: {% endif %}\
+            {{ commit.message | upper_first }}\
+            {% if commit.github.pr_number %} \
+                ([#{{ commit.github.pr_number }}]({{ self::remote_url() }}/pull/{{ commit.github.pr_number }}))\
+            {% endif %}\
+            {%- if commit.github.username %} by [@{{ commit.github.username }}](https://github.com/{{ commit.github.username }}){%- endif %}
+    {%- endfor %}
+{% endfor %}
+
+{%- if github.contributors | filter(attribute="is_first_time", value=true) | length != 0 %}
+
+  ### New Contributors
+{%- for contributor in github.contributors | filter(attribute="is_first_time", value=true) %}
+  * @{{ contributor.username }} made their first contribution
+    {%- if contributor.pr_number %} in \
+      [#{{ contributor.pr_number }}]({{ self::remote_url() }}/pull/{{ contributor.pr_number }}) \
+    {%- endif %}
+{%- endfor %}
+{%- endif -%}
+
+{% if version %}
+    {% if previous.version %}
+      **Full Changelog**: [{{ previous.version }}...{{ version }}]({{ self::remote_url() }}/compare/{{ previous.version }}...{{ version }})
+    {% endif %}
+{% else -%}
+    {% raw %}\n{% endraw %}
+{% endif %}
+"""
+# Remove leading and trailing whitespace
+trim = true
+
+[git]
+# Parse conventional commits
+conventional_commits = true
+filter_unconventional = true
+
+# Process each line of a commit as an individual commit
+split_commits = false
+
+# Regex for preprocessing the commit messages
+commit_preprocessors = [
+  # Remove PR number from message (we add it back via GitHub integration)
+  { pattern = '\(#(\d+)\)', replace = '' },
+  # Strip conventional commit prefix for cleaner messages
+  # (group header already shows the type)
+]
+
+# Regex for parsing and grouping commits
+commit_parsers = [
+  # Skip noise: translation commits from Crowdin
+  { message = "^New translations", skip = true },
+  { message = "^Update source file", skip = true },
+  # Skip merge commits
+  { message = "^Merge", skip = true },
+  # Skip version bump commits
+  { message = "^v\\d+", skip = true },
+  { message = "^chore: update VirusTotal", skip = true },
+
+  # Group by conventional commit type
+  { message = "^feat", group = "<!-- 0 -->New Features" },
+  { message = "^fix", group = "<!-- 1 -->Bug Fixes" },
+  { message = "^perf", group = "<!-- 2 -->Performance" },
+  { message = "^refactor", group = "<!-- 3 -->Refactoring" },
+  { message = "^doc", group = "<!-- 4 -->Documentation" },
+  { message = "^style", group = "<!-- 5 -->Styling" },
+  { message = "^test", group = "<!-- 6 -->Testing" },
+  { message = "^chore\\(deps\\)", group = "<!-- 7 -->Dependencies" },
+  { message = "^chore\\(l10n\\)", skip = true },
+  { message = "^chore|^ci", group = "<!-- 8 -->Chores" },
+]
+
+# Protect breaking changes from being skipped
+protect_breaking_commits = true
+
+# Filter out commits by matching patterns
+filter_commits = false
+
+# Tag pattern for version detection
+tag_pattern = "v[0-9].*"
+
+# Sort commits by newest first
+sort_commits = "newest"
+
+[remote.github]
+owner = "zarzet"
+repo = "SpotiFLAC-Mobile"
diff --git a/lib/providers/local_library_provider.dart b/lib/providers/local_library_provider.dart
index fe9f21ad..1dd10dd7 100644
--- a/lib/providers/local_library_provider.dart
+++ b/lib/providers/local_library_provider.dart
@@ -9,6 +9,7 @@ import 'package:spotiflac_android/services/library_database.dart';
 import 'package:spotiflac_android/services/notification_service.dart';
 import 'package:spotiflac_android/services/platform_bridge.dart';
 import 'package:spotiflac_android/utils/logger.dart';
+import 'package:spotiflac_android/utils/path_match_keys.dart';
 
 final _log = AppLogger('LocalLibrary');
 
@@ -193,74 +194,11 @@ class LocalLibraryNotifier extends Notifier<LocalLibraryState> {
     await _loadFromDatabase();
   }
 
-  Set<String> _buildPathMatchKeys(String? filePath) {
-    final raw = filePath?.trim() ?? '';
-    if (raw.isEmpty) return const {};
-
-    final cleaned = raw.startsWith('EXISTS:') ? raw.substring(7) : raw;
-    final keys = <String>{};
-
-    void addNormalized(String value) {
-      final trimmed = value.trim();
-      if (trimmed.isEmpty) return;
-      keys.add(trimmed);
-      keys.add(trimmed.toLowerCase());
-      if (trimmed.contains('\\')) {
-        final slash = trimmed.replaceAll('\\', '/');
-        keys.add(slash);
-        keys.add(slash.toLowerCase());
-      }
-      if (trimmed.contains('%')) {
-        try {
-          final decoded = Uri.decodeFull(trimmed);
-          keys.add(decoded);
-          keys.add(decoded.toLowerCase());
-        } catch (_) {}
-      }
-
-      Uri? parsed;
-      try {
-        parsed = Uri.parse(trimmed);
-      } catch (_) {}
-
-      if (parsed != null && parsed.hasScheme) {
-        final noQueryOrFragment = parsed.replace(query: null, fragment: null);
-        keys.add(noQueryOrFragment.toString());
-        keys.add(noQueryOrFragment.toString().toLowerCase());
-
-        if (parsed.scheme == 'file') {
-          try {
-            final fileOnly = parsed.toFilePath();
-            if (fileOnly.isNotEmpty) {
-              keys.add(fileOnly);
-              keys.add(fileOnly.toLowerCase());
-              if (fileOnly.contains('\\')) {
-                final slash = fileOnly.replaceAll('\\', '/');
-                keys.add(slash);
-                keys.add(slash.toLowerCase());
-              }
-            }
-          } catch (_) {}
-        }
-      } else if (trimmed.startsWith('/')) {
-        try {
-          final asFileUri = Uri.file(trimmed).toString();
-          keys.add(asFileUri);
-          keys.add(asFileUri.toLowerCase());
-        } catch (_) {}
-      }
-    }
-
-    addNormalized(cleaned);
-
-    return keys;
-  }
-
   bool _isDownloadedPath(String? filePath, Set<String> downloadedPathKeys) {
     if (filePath == null || filePath.isEmpty || downloadedPathKeys.isEmpty) {
       return false;
     }
-    final candidateKeys = _buildPathMatchKeys(filePath);
+    final candidateKeys = buildPathMatchKeys(filePath);
     for (final key in candidateKeys) {
       if (downloadedPathKeys.contains(key)) {
         return true;
@@ -322,8 +260,9 @@ class LocalLibraryNotifier extends Notifier<LocalLibraryState> {
     String? resolvedPath;
     bool didStartSecurityAccess = false;
     if (Platform.isIOS && iosBookmark != null && iosBookmark.isNotEmpty) {
-      resolvedPath =
-          await PlatformBridge.startAccessingIosBookmark(iosBookmark);
+      resolvedPath = await PlatformBridge.startAccessingIosBookmark(
+        iosBookmark,
+      );
       if (resolvedPath != null) {
         didStartSecurityAccess = true;
         _log.i('Started iOS security-scoped access: $resolvedPath');
@@ -354,7 +293,7 @@ class LocalLibraryNotifier extends Notifier<LocalLibraryState> {
       };
       final downloadedPathKeys = <String>{};
       for (final path in allHistoryPaths) {
-        downloadedPathKeys.addAll(_buildPathMatchKeys(path));
+        downloadedPathKeys.addAll(buildPathMatchKeys(path));
       }
       _log.i(
         'Excluding ${allHistoryPaths.length} downloaded files from library scan '
@@ -834,8 +773,9 @@ class LocalLibraryNotifier extends Notifier<LocalLibraryState> {
   Future<int> cleanupMissingFiles({String? iosBookmark}) async {
     bool didStartSecurityAccess = false;
     if (Platform.isIOS && iosBookmark != null && iosBookmark.isNotEmpty) {
-      final resolved =
-          await PlatformBridge.startAccessingIosBookmark(iosBookmark);
+      final resolved = await PlatformBridge.startAccessingIosBookmark(
+        iosBookmark,
+      );
       if (resolved != null) {
         didStartSecurityAccess = true;
       }
diff --git a/lib/screens/queue_tab.dart b/lib/screens/queue_tab.dart
index 1aa3bad3..e05429c9 100644
--- a/lib/screens/queue_tab.dart
+++ b/lib/screens/queue_tab.dart
@@ -29,6 +29,7 @@ import 'package:spotiflac_android/screens/downloaded_album_screen.dart';
 import 'package:spotiflac_android/screens/library_tracks_folder_screen.dart';
 import 'package:spotiflac_android/screens/local_album_screen.dart';
 import 'package:spotiflac_android/utils/clickable_metadata.dart';
+import 'package:spotiflac_android/utils/path_match_keys.dart';
 import 'package:spotiflac_android/utils/string_utils.dart';
 
 enum LibraryItemSource { downloaded, local }
@@ -2112,10 +2113,22 @@ class _QueueTabState extends ConsumerState<QueueTab> {
       if (count > 1) albumCount++;
     }
 
+    final downloadedPathKeys = <String>{};
+    for (final item in items) {
+      downloadedPathKeys.addAll(buildPathMatchKeys(item.filePath));
+    }
+
+    final dedupedLocalItems = localItems
+        .where((item) {
+          final localPathKeys = buildPathMatchKeys(item.filePath);
+          return !localPathKeys.any(downloadedPathKeys.contains);
+        })
+        .toList(growable: false);
+
     // Calculate local library stats
     final localAlbumCounts = <String, int>{};
     final localAlbumMap = <String, List<LocalLibraryItem>>{};
-    for (final item in localItems) {
+    for (final item in dedupedLocalItems) {
       final key =
           '${item.albumName.toLowerCase()}|${(item.albumArtist ?? item.artistName).toLowerCase()}';
       localAlbumCounts[key] = (localAlbumCounts[key] ?? 0) + 1;
@@ -2841,8 +2854,24 @@ class _QueueTabState extends ConsumerState<QueueTab> {
         .map((item) => UnifiedLibraryItem.fromLocalLibrary(item))
         .toList(growable: false);
 
-    final merged = <UnifiedLibraryItem>[...unifiedDownloaded, ...unifiedLocal]
-      ..sort((a, b) => b.addedAt.compareTo(a.addedAt));
+    final downloadedPathKeys = <String>{};
+    for (final item in unifiedDownloaded) {
+      downloadedPathKeys.addAll(buildPathMatchKeys(item.filePath));
+    }
+
+    final dedupedUnifiedLocal = <UnifiedLibraryItem>[];
+    for (final item in unifiedLocal) {
+      final localPathKeys = buildPathMatchKeys(item.filePath);
+      final overlapsDownloaded = localPathKeys.any(downloadedPathKeys.contains);
+      if (!overlapsDownloaded) {
+        dedupedUnifiedLocal.add(item);
+      }
+    }
+
+    final merged = <UnifiedLibraryItem>[
+      ...unifiedDownloaded,
+      ...dedupedUnifiedLocal,
+    ]..sort((a, b) => b.addedAt.compareTo(a.addedAt));
 
     _unifiedItemsCache[filterMode] = _UnifiedCacheEntry(
       historyItems: historyItems,
diff --git a/lib/utils/path_match_keys.dart b/lib/utils/path_match_keys.dart
new file mode 100644
index 00000000..ace140dd
--- /dev/null
+++ b/lib/utils/path_match_keys.dart
@@ -0,0 +1,104 @@
+import 'dart:io';
+
+const _androidStoragePathAliases = <String>[
+  '/storage/emulated/0',
+  '/storage/emulated/legacy',
+  '/storage/self/primary',
+  '/sdcard',
+  '/mnt/sdcard',
+];
+
+Set<String> buildPathMatchKeys(String? filePath) {
+  final raw = filePath?.trim() ?? '';
+  if (raw.isEmpty) return const {};
+
+  final cleaned = raw.startsWith('EXISTS:') ? raw.substring(7).trim() : raw;
+  if (cleaned.isEmpty) return const {};
+
+  final keys = <String>{};
+  final visited = <String>{};
+
+  void addNormalized(String value) {
+    final trimmed = value.trim();
+    if (trimmed.isEmpty) return;
+    if (!visited.add(trimmed)) return;
+
+    keys.add(trimmed);
+    keys.add(trimmed.toLowerCase());
+
+    if (trimmed.contains('\\')) {
+      final slash = trimmed.replaceAll('\\', '/');
+      if (slash != trimmed) {
+        addNormalized(slash);
+      }
+    }
+
+    if (trimmed.contains('%')) {
+      try {
+        final decoded = Uri.decodeFull(trimmed);
+        if (decoded != trimmed) {
+          addNormalized(decoded);
+        }
+      } catch (_) {}
+    }
+
+    Uri? parsed;
+    try {
+      parsed = Uri.parse(trimmed);
+    } catch (_) {}
+
+    if (parsed != null && parsed.hasScheme) {
+      final withoutQueryOrFragment = parsed.replace(
+        query: null,
+        fragment: null,
+      );
+      final uriString = withoutQueryOrFragment.toString();
+      keys.add(uriString);
+      keys.add(uriString.toLowerCase());
+
+      if (parsed.scheme == 'file') {
+        try {
+          addNormalized(parsed.toFilePath());
+        } catch (_) {}
+      }
+    } else if (trimmed.startsWith('/')) {
+      try {
+        final asFileUri = Uri.file(trimmed).toString();
+        keys.add(asFileUri);
+        keys.add(asFileUri.toLowerCase());
+      } catch (_) {}
+    }
+
+    if (Platform.isAndroid) {
+      for (final alias in _androidEquivalentPaths(trimmed)) {
+        if (alias != trimmed) {
+          addNormalized(alias);
+        }
+      }
+    }
+  }
+
+  addNormalized(cleaned);
+  return keys;
+}
+
+Iterable<String> _androidEquivalentPaths(String path) {
+  final normalized = path.replaceAll('\\', '/');
+  final lower = normalized.toLowerCase();
+  String? suffix;
+
+  for (final prefix in _androidStoragePathAliases) {
+    if (lower == prefix) {
+      suffix = '';
+      break;
+    }
+    final withSlash = '$prefix/';
+    if (lower.startsWith(withSlash)) {
+      suffix = normalized.substring(prefix.length);
+      break;
+    }
+  }
+
+  if (suffix == null) return const [];
+  return _androidStoragePathAliases.map((prefix) => '$prefix$suffix');
+}