diff --git a/src/utils/normalize.test.ts b/src/utils/normalize.test.ts
index 40e7ac7..5f825a6 100644
--- a/src/utils/normalize.test.ts
+++ b/src/utils/normalize.test.ts
@@ -54,6 +54,32 @@ describe('utils/normalize.ts', () => {
     expect(normalizeRomanText('Seibuen')).toBe('Seibuen');
   });
 
+  it('replaces Mine with a spelling English TTS reads as みね', () => {
+    // 英語 TTS は "Mine" を英単語の「まいん」と読むため、辞書語の綴りへ倒す
+    expect(normalizeRomanText('Change here for the Mine Line.')).toBe(
+      'Change here for the Me-nay Line.'
+    );
+    expect(normalizeRomanText('The next station is MINE.')).toBe(
+      'The next station is Me-nay.'
+    );
+    expect(normalizeRomanText('The next station is Nishi-Mine.')).toBe(
+      'The next station is Nishi-me-nay.'
+    );
+    // 宇都宮ライトレールの峰（Mine）のような単独の駅名
+    expect(normalizeRomanText('The next stop is Mine.')).toBe(
+      'The next stop is Me-nay.'
+    );
+    expect(normalizeRomanText('This train is bound for Mine.')).toBe(
+      'This train is bound for Me-nay.'
+    );
+    // 語中・語頭に埋まった mine も置換する
+    expect(normalizeRomanText('Takamine')).toBe('Taka-me-nay');
+    expect(normalizeRomanText('Minezaki')).toBe('Me-nay-zaki');
+    expect(normalizeRomanText('KAMINEYAMA')).toBe('Ka-me-nay-yama');
+    // mine を含まない語は変えない
+    expect(normalizeRomanText('Minami-Urawa')).toBe('Minami-urawa');
+  });
+
   it('keeps Kay-say stable when normalized twice', () => {
     // 二重に適用しても結果が変わらないこと（キャッシュキーの安定性）
     expect(normalizeRomanText(normalizeRomanText('Keisei Main Line'))).toBe(
@@ -62,6 +88,12 @@ describe('utils/normalize.ts', () => {
     expect(normalizeRomanText(normalizeRomanText('Seibu Shinjuku Line'))).toBe(
       'Say-boo Shinjuku Line'
     );
+    expect(normalizeRomanText(normalizeRomanText('Mine Line'))).toBe(
+      'Me-nay Line'
+    );
+    expect(normalizeRomanText(normalizeRomanText('Takamine'))).toBe(
+      'Taka-me-nay'
+    );
   });
 
   it.each(['Tokyo', 'tOkyo'])('text: %s', (text) => {
diff --git a/src/utils/normalize.ts b/src/utils/normalize.ts
index cfd4deb..6c44efc 100644
--- a/src/utils/normalize.ts
+++ b/src/utils/normalize.ts
@@ -15,6 +15,16 @@ const capitalizeSegment = (seg: string): string => {
     : seg;
 };
 
+// Takamine / Minezaki のように語中・語頭に埋まった "mine" も読み替える。
+// 前後に英字が続く場合はハイフンで区切って "Me-nay" を独立させ、
+// 語頭のときだけ先頭を大文字にする（Takamine → Taka-me-nay）
+const replaceMine = (match: string, offset: number, str: string): string => {
+  const before = offset > 0 && /[A-Za-z]/.test(str.charAt(offset - 1));
+  const after = /[A-Za-z]/.test(str.charAt(offset + match.length));
+  const head = /[A-Z]/.test(match.charAt(0)) ? 'Me' : 'me';
+  return `${before ? '-' : ''}${head}-nay${after ? '-' : ''}`;
+};
+
 // テキストノード（SSML タグの外側）だけに掛ける正規化。タグやその属性値
 // （<sub alias="Sta."> や <phoneme ph="..."> 等）を壊さないため、タグ部分には適用しない。
 const normalizeTextNode = (text: string): string =>
@@ -43,6 +53,9 @@ const normalizeTextNode = (text: string): string =>
     // ハイフン連結の駅名も語単位で置換する
     .replace(/\bKeisei\b/gi, 'Kay-say')
     .replace(/\bSeibu\b/gi, 'Say-boo')
+    // 「mine（美祢・峰など）」は英語 TTS が英単語 "mine"（まいん）として読むため、
+    // 同じく辞書語 "Me" + "nay" の連結で /mi.neɪ/（みね）に寄せる
+    .replace(/mine/gi, replaceMine)
     // 都営バスを想定
     .replace(/\bSta\./gi, ' Station')
     .replace(/\bUniv\./gi, ' University')