diff --git a/package-lock.json b/package-lock.json
index d9ebd95..7409985 100644
--- a/package-lock.json
+++ b/package-lock.json
@@ -1,12 +1,12 @@
 {
   "name": "speechmarkdown-js",
-  "version": "2.1.1",
+  "version": "2.3.1",
   "lockfileVersion": 3,
   "requires": true,
   "packages": {
     "": {
       "name": "speechmarkdown-js",
-      "version": "2.1.1",
+      "version": "2.3.1",
       "license": "MIT",
       "dependencies": {
         "myna-parser": "^2.5.1",
diff --git a/src/SpeechMarkdownGrammar.ts b/src/SpeechMarkdownGrammar.ts
index cb32d05..2bde628 100644
--- a/src/SpeechMarkdownGrammar.ts
+++ b/src/SpeechMarkdownGrammar.ts
@@ -87,6 +87,64 @@ export function speechMarkdownGrammar(myna: any): any {
     this.time = m.seq(this.number, this.timeUnit).ast;
     this.shortBreak = m.seq('[', this.time, ']').ast;
 
+    // Expressive audio tags: [laugh], [sigh], [cough], etc.
+    this.expressiveValue = m.keywords(
+      'laugh',
+      'laughter',
+      'sigh',
+      'cough',
+      'cheer',
+      'cheering',
+      'cry',
+      'crying',
+      'gasp',
+      'groan',
+      'groaning',
+      'hum',
+      'hmm',
+      'mm-hmm',
+      'oh',
+      'sniff',
+      'whew',
+      'wow',
+      'yawn',
+      'yeah',
+      'huh',
+      'tsk',
+      'uh-huh',
+      'mmm',
+      'mhm',
+      'ahem',
+      'applause',
+      'boo',
+      'giggle',
+      'hiccup',
+      'hurray',
+      'moan',
+      'pant',
+      'scream',
+      'shush',
+      'sneeze',
+      'throat-clear',
+      'wheeze',
+      'whimper',
+      'yay',
+      'bleh',
+      'eek',
+      'hmm',
+      'huh',
+      'meh',
+      'ooh',
+      'pfft',
+      'phew',
+      'psst',
+      'shh',
+      'tsk-tsk',
+      'uh-oh',
+      'umph',
+    ).ast;
+    this.expressive = m.seq('[', this.expressiveValue, ']').ast;
+
     // this.break = m.seq('[break:', this.time , ']').ast;
 
     // this.string = m.doubleQuoted(this.quoteChar.zeroOrMore).ast;
@@ -449,6 +507,7 @@ export function speechMarkdownGrammar(myna: any): any {
         this.shortSub,
         this.textModifier,
         this.emphasis,
+        this.expressive,
         this.shortBreak,
         this.break,
         this.audio,
diff --git a/src/formatters/ElevenLabsFormatter.ts b/src/formatters/ElevenLabsFormatter.ts
index 5b0181d..d9dd2fc 100644
--- a/src/formatters/ElevenLabsFormatter.ts
+++ b/src/formatters/ElevenLabsFormatter.ts
@@ -116,6 +116,11 @@ export class ElevenLabsFormatter extends SsmlFormatterBase {
         const time = ast.children[0].allText;
         return this.addTagWithAttrs(lines, null, 'break', { time: time });
       }
+      case 'expressive': {
+        const value = ast.children[0].allText;
+        lines.push(`[${value}]`);
+        return lines;
+      }
       case 'break': {
         const val = ast.children[0].allText;
         let time = val;
diff --git a/src/formatters/TextFormatter.ts b/src/formatters/TextFormatter.ts
index c41c5a8..11c2fe2 100644
--- a/src/formatters/TextFormatter.ts
+++ b/src/formatters/TextFormatter.ts
@@ -83,6 +83,12 @@ export class TextFormatter extends FormatterBase {
       case 'audio':
         return lines;
 
+      case 'expressive': {
+        const value = ast.children[0].allText;
+        lines.push(`[${value}]`);
+        return lines;
+      }
+
       default: {
         this.processAst(ast.children, lines);
         return lines;
diff --git a/src/formatters/W3cSsmlFormatter.ts b/src/formatters/W3cSsmlFormatter.ts
index d4f2e87..a270a78 100644
--- a/src/formatters/W3cSsmlFormatter.ts
+++ b/src/formatters/W3cSsmlFormatter.ts
@@ -198,6 +198,11 @@ export class W3cSsmlFormatter extends SsmlFormatterBase {
         const time = ast.children[0].allText;
         return this.addTagWithAttrs(lines, null, 'break', { time });
       }
+      case 'expressive': {
+        const value = ast.children[0].allText;
+        lines.push(`[${value}]`);
+        return lines;
+      }
       case 'break': {
         const val = ast.children[0].allText;
         let attrs = {};
diff --git a/tests/expressive.spec.ts b/tests/expressive.spec.ts
new file mode 100644
index 0000000..6902e6c
--- /dev/null
+++ b/tests/expressive.spec.ts
@@ -0,0 +1,111 @@
+import dedent from 'ts-dedent';
+import { SpeechMarkdown } from '../src/SpeechMarkdown';
+
+describe('expressive', () => {
+  const speech = new SpeechMarkdown();
+
+  const markdown = dedent`
+    Hello [laugh] world
+  `;
+
+  test('converts to SSML - W3C', () => {
+    const ssml = speech.toSSML(markdown, { platform: 'w3c' });
+    const expected = dedent`
+      <speak>
+      Hello [laugh] world
+      </speak>
+    `;
+    expect(ssml).toBe(expected);
+  });
+
+  test('converts to SSML - ElevenLabs', () => {
+    const ssml = speech.toSSML(markdown, { platform: 'elevenlabs' });
+    expect(ssml).toBe('Hello [laugh] world');
+  });
+
+  test('converts to Plain Text', () => {
+    const text = speech.toText(markdown);
+    expect(text).toBe('Hello [laugh] world');
+  });
+
+  test('converts to SSML - Amazon Polly (strips)', () => {
+    const ssml = speech.toSSML(markdown, { platform: 'amazon-polly' });
+    const expected = dedent`
+      <speak>
+      Hello  world
+      </speak>
+    `;
+    expect(ssml).toBe(expected);
+  });
+
+  test('converts to SSML - Google Assistant (strips)', () => {
+    const ssml = speech.toSSML(markdown, { platform: 'google-assistant' });
+    const expected = dedent`
+      <speak>
+      Hello  world
+      </speak>
+    `;
+    expect(ssml).toBe(expected);
+  });
+
+  test('converts to SSML - Microsoft Azure (strips)', () => {
+    const ssml = speech.toSSML(markdown, { platform: 'microsoft-azure' });
+    const expected = dedent`
+      <speak>
+      Hello  world
+      </speak>
+    `;
+    expect(ssml).toBe(expected);
+  });
+});
+
+describe('expressive multiple', () => {
+  const speech = new SpeechMarkdown();
+
+  const markdown = dedent`
+    Hello [laugh] how are you [sigh] I'm fine [cough]
+  `;
+
+  test('converts to SSML - W3C', () => {
+    const ssml = speech.toSSML(markdown, { platform: 'w3c' });
+    const expected = dedent`
+      <speak>
+      Hello [laugh] how are you [sigh] I'm fine [cough]
+      </speak>
+    `;
+    expect(ssml).toBe(expected);
+  });
+
+  test('converts to Plain Text', () => {
+    const text = speech.toText(markdown);
+    expect(text).toBe("Hello [laugh] how are you [sigh] I'm fine [cough]");
+  });
+});
+
+describe('expressive edge cases', () => {
+  const speech = new SpeechMarkdown();
+
+  test('expressive does not conflict with shortBreak [250ms]', () => {
+    const md = 'Hello [laugh] wait [250ms] world';
+    const ssml = speech.toSSML(md, { platform: 'w3c' });
+    const expected = dedent`
+      <speak>
+      Hello [laugh] wait <break time="250ms"/> world
+      </speak>
+    `;
+    expect(ssml).toBe(expected);
+  });
+
+  test('expressive does not conflict with textModifier', () => {
+    const md = '(hello)[emphasis:"strong"] [laugh]';
+    const ssml = speech.toSSML(md, { platform: 'w3c' });
+    expect(ssml).toContain('[laugh]');
+    expect(ssml).toContain('<emphasis');
+  });
+
+  test('unknown bracket content falls through as plain text', () => {
+    const md = 'Hello [unknownthing] world';
+    const text = speech.toText(md);
+    expect(text).toBe('Hello [unknownthing] world');
+  });
+});