diff --git a/src/tokenizer.ts b/src/tokenizer.ts
index 4f93a03..cc2f55c 100644
--- a/src/tokenizer.ts
+++ b/src/tokenizer.ts
@@ -725,7 +725,7 @@ export class Tokenizer {
             endReg.lastIndex = 0;
 
             // Clip maskedSrc to same section of string as src (move to lexer?)
-            maskedSrc = maskedSrc.slice(-1 * src.length + match[0].length - 1);
+            maskedSrc = maskedSrc.slice(-1 * src.length + lLength);
 
             while ((match = endReg.exec(maskedSrc)) != null) {
                 // eslint-disable-next-line
@@ -755,8 +755,9 @@ export class Tokenizer {
 
                 // Remove extra characters. *a*** -> *a*
                 rLength = Math.min(rLength, rLength + delimTotal + midDelimTotal);
-
-                const raw = [...src].slice(0, lLength + match.index + rLength + 1).join("");
+                // char length can be >1 for unicode characters;
+                const lastCharLength = [...match[0]]![0]!.length;
+                const raw = src.slice(0, lLength + match.index + lastCharLength + rLength);
 
                 // Create `em` if smallest delimiter has odd char count. *a***
                 if (Math.min(lLength, rLength) % 2) {