diff --git a/CHANGELOG.md b/CHANGELOG.md
index 561de76..ebf37ca 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,6 +6,12 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 
+## [2.0.4] - WIP
+### Fixed
+
+- Use TextEncoder and TextDecoder if available, #228.
+
+
 ## [2.0.3] - 2021-01-09
 ### Fixed
 
@@ -174,6 +180,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - First release.
 
 
+[2.0.4]: https://github.com/nodeca/pako/compare/2.0.3...2.0.4
 [2.0.3]: https://github.com/nodeca/pako/compare/2.0.2...2.0.3
 [2.0.2]: https://github.com/nodeca/pako/compare/2.0.1...2.0.2
 [2.0.1]: https://github.com/nodeca/pako/compare/2.0.0...2.0.1
diff --git a/lib/utils/strings.js b/lib/utils/strings.js
index b48515e..c8f097c 100644
--- a/lib/utils/strings.js
+++ b/lib/utils/strings.js
@@ -24,6 +24,10 @@ _utf8len[254] = _utf8len[254] = 1; // Invalid sequence start
 
 // convert string to array (typed, when possible)
 module.exports.string2buf = (str) => {
+  if (typeof TextEncoder === 'function' && TextEncoder.prototype.encode) {
+    return new TextEncoder().encode(str);
+  }
+
   let buf, c, c2, m_pos, i, str_len = str.length, buf_len = 0;
 
   // count binary size
@@ -97,9 +101,14 @@ const buf2binstring = (buf, len) => {
 
 // convert array to string
 module.exports.buf2string = (buf, max) => {
-  let i, out;
   const len = max || buf.length;
 
+  if (typeof TextDecoder === 'function' && TextDecoder.prototype.decode) {
+    return new TextDecoder().decode(buf.subarray(0, max));
+  }
+
+  let i, out;
+
   // Reserve max possible length (2 words per char)
   // NB: by unknown reasons, Array is significantly faster for
   //     String.fromCharCode.apply than Uint16Array.
diff --git a/test/strings.js b/test/strings.js
index 9d9cf5c..41e2b9f 100644
--- a/test/strings.js
+++ b/test/strings.js
@@ -40,6 +40,19 @@ describe('Encode/Decode', () => {
   // use node Buffer internal conversion as "done right"
   const utf8sample = new Uint8Array(Buffer.from(utf16sample));
 
+  let _TextEncoder, _TextDecoder;
+
+  /* eslint-disable no-global-assign, no-native-reassign */
+  beforeEach(() => {
+    _TextEncoder = TextEncoder;
+    _TextDecoder = TextDecoder;
+  });
+
+  afterEach(() => {
+    TextEncoder = _TextEncoder;
+    TextDecoder = _TextDecoder;
+  });
+
   it('utf-8 border detect', () => {
     const ub = strings.utf8border;
     assert.strictEqual(ub(utf8sample, 1), 1);
@@ -75,10 +88,19 @@ describe('Encode/Decode', () => {
       strings.string2buf(utf16sample),
       utf8sample
     );
+
+    TextEncoder = null;
+    assert.deepStrictEqual(
+      strings.string2buf(utf16sample),
+      utf8sample
+    );
   });
 
   it('Decode utf8 buf to string', () => {
     assert.ok(strings.buf2string(utf8sample), utf16sample);
+
+    TextDecoder = null;
+    assert.ok(strings.buf2string(utf8sample), utf16sample);
   });
 
 });