diff --git a/soundoutput.py b/soundoutput.py
index c19df4e..3b30ea1 100644
--- a/soundoutput.py
+++ b/soundoutput.py
@@ -70,6 +70,9 @@ def send_audio(self):
                 self.pcm = self.pcm[samples:]
                 self.lock.release()
                 
+                if len(to_encode) != samples:  # pad to_encode if needed to match sample length
+                    to_encode += '\x00'*(samples-len(to_encode))
+
                 encoded = self.encoder.encode(to_encode)
                          
                 audio_encoded += self.encoder_framesize