diff --git a/font_collector/font.py b/font_collector/font.py
index 3cfba26..a147033 100644
--- a/font_collector/font.py
+++ b/font_collector/font.py
@@ -302,6 +302,16 @@ def get_missing_glyphs(self, text: Sequence[str]) -> Set[str]:
                 if cmap_encoding == "unicode":
                     codepoint = ord(char)
                 else:
+                    if cmap_encoding == "unknown" and platform_id == 3 and encoding_id == 0:
+                        cmap_encoding = FontParser.get_symbol_cmap_encoding(face)
+
+                        if cmap_encoding is None:
+                            # Fallback if guess fails
+                            cmap_encoding = "cp1252"
+                    else:
+                        # cmap not supported 
+                        continue
+
                     try:
                         codepoint = int.from_bytes(char.encode(cmap_encoding), "big")
                     except UnicodeEncodeError:
diff --git a/font_collector/font_parser.py b/font_collector/font_parser.py
index 6ee4be4..4f54f08 100644
--- a/font_collector/font_parser.py
+++ b/font_collector/font_parser.py
@@ -52,7 +52,7 @@ class FontParser:
             0: "mac_roman",
         },
         3: {  # Microsoft
-            0: "unicode",
+            0: "unknown",
             1: "unicode",
             2: "cp932",
             3: "cp936",