diff --git a/native_client/client.cc b/native_client/client.cc
index d2dc6085..04ba452e 100644
--- a/native_client/client.cc
+++ b/native_client/client.cc
@@ -246,7 +246,7 @@ ProcessFile(ModelState* context, const char* path, bool show_times)
 
   if (result.string) {
     printf("%s\n", result.string);
-    free((void*)result.string);
+    DS_FreeString((char*)result.string);
   }
 
   if (show_times) {
diff --git a/native_client/deepspeech.cc b/native_client/deepspeech.cc
index 344c301d..fc0b2004 100644
--- a/native_client/deepspeech.cc
+++ b/native_client/deepspeech.cc
@@ -926,6 +926,12 @@ DS_FreeMetadata(Metadata* m)
   }
 }
 
+void
+DS_FreeString(char* str)
+{
+  free(str);
+}
+
 void
 DS_PrintVersions() {
   std::cerr << "TensorFlow: " << tf_local_git_version() << std::endl;
diff --git a/native_client/deepspeech.h b/native_client/deepspeech.h
index 620d2558..7dba202a 100644
--- a/native_client/deepspeech.h
+++ b/native_client/deepspeech.h
@@ -116,8 +116,8 @@ int DS_EnableDecoderWithLM(ModelState* aCtx,
  * @param aBufferSize The number of samples in the audio signal.
  * @param aSampleRate The sample-rate of the audio signal.
  *
- * @return The STT result. The user is responsible for freeing the string.
- *         Returns NULL on error.
+ * @return The STT result. The user is responsible for freeing the string using
+ *         {@link DS_FreeString()}. Returns NULL on error.
  */
 DEEPSPEECH_EXPORT
 char* DS_SpeechToText(ModelState* aCtx,
@@ -187,7 +187,7 @@ void DS_FeedAudioContent(StreamingState* aSctx,
  * @param aSctx A streaming state pointer returned by {@link DS_SetupStream()}.
  *
  * @return The STT intermediate result. The user is responsible for freeing the
- *         string.
+ *         string using {@link DS_FreeString()}.
  */
 DEEPSPEECH_EXPORT
 char* DS_IntermediateDecode(StreamingState* aSctx);
@@ -198,7 +198,8 @@ char* DS_IntermediateDecode(StreamingState* aSctx);
  *
  * @param aSctx A streaming state pointer returned by {@link DS_SetupStream()}.
  *
- * @return The STT result. The user is responsible for freeing the string.
+ * @return The STT result. The user is responsible for freeing the string using
+ *         {@link DS_FreeString()}.
  *
  * @note This method will free the state pointer (@p aSctx).
  */
@@ -235,7 +236,13 @@ void DS_DiscardStream(StreamingState* aSctx);
  * @brief Free memory allocated for metadata information.
  */
 DEEPSPEECH_EXPORT
-void DS_FreeMetadata(Metadata* m); 
+void DS_FreeMetadata(Metadata* m);
+
+/**
+ * @brief Free a char* string returned by the DeepSpeech API.
+ */
+DEEPSPEECH_EXPORT
+void DS_FreeString(char* str);
 
 /**
  * @brief Print version of this library and of the linked TensorFlow library.
diff --git a/native_client/java/jni/deepspeech.i b/native_client/java/jni/deepspeech.i
index fb7caac1..50a65bf8 100644
--- a/native_client/java/jni/deepspeech.i
+++ b/native_client/java/jni/deepspeech.i
@@ -15,6 +15,11 @@
 %pointer_functions(ModelState*, modelstatep);
 %pointer_functions(StreamingState*, streamingstatep);
 
+%typemap(newfree) char* "DS_FreeString($1);";
+%newobject DS_SpeechToText;
+%newobject DS_IntermediateDecode;
+%newobject DS_FinishStream;
+
 %rename ("%(strip:[DS_])s") "";
 
 %include "../deepspeech.h"
diff --git a/native_client/javascript/deepspeech.i b/native_client/javascript/deepspeech.i
index 733465e4..a2a9e148 100644
--- a/native_client/javascript/deepspeech.i
+++ b/native_client/javascript/deepspeech.i
@@ -27,7 +27,7 @@ using namespace node;
 
 
 // make sure the string returned by SpeechToText is freed
-%typemap(newfree) char* "free($1);";
+%typemap(newfree) char* "DS_FreeString($1);";
 %newobject DS_SpeechToText;
 %newobject DS_IntermediateDecode;
 %newobject DS_FinishStream;
diff --git a/native_client/python/impl.i b/native_client/python/impl.i
index 03d08710..1617c029 100644
--- a/native_client/python/impl.i
+++ b/native_client/python/impl.i
@@ -33,7 +33,7 @@ import_array();
   %append_output(SWIG_NewPointerObj(%as_voidptr(*$1), $*1_descriptor, 0));
 }
 
-%typemap(newfree) char* "free($1);";
+%typemap(newfree) char* "DS_FreeString($1);";
 %newobject DS_SpeechToText;
 %newobject DS_IntermediateDecode;
 %newobject DS_FinishStream;