diff --git a/tensorflow/lite/micro/examples/micro_speech/osx/audio_provider.cc b/tensorflow/lite/micro/examples/micro_speech/osx/audio_provider.cc
index cdb7a33b1b2..af16d2aeeb2 100644
--- a/tensorflow/lite/micro/examples/micro_speech/osx/audio_provider.cc
+++ b/tensorflow/lite/micro/examples/micro_speech/osx/audio_provider.cc
@@ -61,10 +61,7 @@ void OnAudioBufferFilledCallback(
 // device on MacOS.
 TfLiteStatus InitAudioRecording(tflite::ErrorReporter* error_reporter) {
   // Set up the format of the audio - single channel, 32-bit float at 16KHz.
-  #pragma clang diagnostic push
-  #pragma clang diagnostic ignored "-Wmissing-field-initializers"
-  AudioStreamBasicDescription recordFormat = {0};
-  #pragma clang diagnostic pop 
+  AudioStreamBasicDescription recordFormat = { };
   recordFormat.mSampleRate = kAudioSampleFrequency;
   recordFormat.mFormatID = kAudioFormatLinearPCM;
   recordFormat.mFormatFlags =