diff --git a/Makefile.am b/Makefile.am
index 7ebef6c5..acce0ba1 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -252,6 +252,7 @@ phsource/phonemes.stamp: \
 	phsource/ph_icelandic \
 	phsource/ph_indonesian \
 	phsource/ph_interlingua \
+	phsource/ph_ipa \
 	phsource/ph_irish \
 	phsource/ph_italian \
 	phsource/ph_japanese \
diff --git a/README.md b/README.md
index 9addf72b..abaa1771 100644
--- a/README.md
+++ b/README.md
@@ -28,7 +28,7 @@
 ----------
 
 The eSpeak NG (Next Generation) Text-to-Speech program is an open source speech
-synthesizer that supports [95 languages and accents](docs/languages.md).
+synthesizer that supports [96 languages and accents](docs/languages.md).
 It is based on the eSpeak engine created by Jonathan Duddington. It uses
 spectral formant synthesis by default which sounds robotic, but can be
 configured to use Klatt formant synthesis or MBROLA to give it a more natural
diff --git a/docs/languages.md b/docs/languages.md
index 47203e61..73e5985a 100644
--- a/docs/languages.md
+++ b/docs/languages.md
@@ -83,6 +83,7 @@ The supported languages are:
 | `inc`       | `or`              | Indic                 | Oriya                       |                        |
 | `cus`       | `om`              | Cushitic              | Oromo                       |                        |
 | `roa`       | `pap`             | Romance               | Papiamento                  |                        |
+| `und`       | `und-fonipa`      |                       | Phonemes<sup>[5]</sup>      | IPA                    |
 | `zls`       | `pl`              | South Slavic          | Polish                      |                        |
 | `roa`       | `pt-BR`           | Romance               | Portuguese                  | Brazillian             |
 | `roa`       | `pt`              | Romance               | Portuguese                  | Portugal               |
@@ -115,3 +116,5 @@ The supported languages are:
 [3] Farsi/Persian written using English (Latin) characters.
 
 [4] Currently, only Hiragana and Katakana are supported.
+
+[5] Currently only supported using the `[[...]]` phoneme blocks.
\ No newline at end of file
diff --git a/espeak-ng-data/voices/und/und-fonipa b/espeak-ng-data/voices/und/und-fonipa
new file mode 100644
index 00000000..b1d63645
--- /dev/null
+++ b/espeak-ng-data/voices/und/und-fonipa
@@ -0,0 +1,4 @@
+name ipa
+language ipa
+
+status testing
diff --git a/phsource/ph_ipa b/phsource/ph_ipa
new file mode 100644
index 00000000..8e3b9141
--- /dev/null
+++ b/phsource/ph_ipa
@@ -0,0 +1,3 @@
+// International Phonetic Alphabet phonemes using the Conlang X-SAMPA (CXS)
+// transcription scheme.
+
diff --git a/phsource/phonemes b/phsource/phonemes
index 3ea7170c..a066b211 100644
--- a/phsource/phonemes
+++ b/phsource/phonemes
@@ -1666,6 +1666,9 @@ endphoneme
 // ADDITIONAL PHONEME TABLES
 //*******************************************************************
 
+phonemetable ipa _
+include ph_ipa
+
 phonemetable jp base
 include ph_japanese