Try Dummy speaker again

supersensor.yaml

@@ -379,6 +379,26 @@ microphone:
     pdm: false
     channel: left
 
+speaker:
+  - platform: i2s_audio
+    id: dummy_speaker
+    i2s_dout_pin: GPIO24
+    dac_type: external
+    bits_per_sample: 16bit
+    sample_rate: 16000
+    channel: mono #mono will have bad performance, however, we are not using the on device speaker
+    buffer_duration: 60ms
+
+media_player:
+  - platform: speaker
+    name: None
+    id: dummy_media_player
+    announcement_pipeline:
+      speaker: dummy_speaker
+      format: WAV
+    codec_support_enabled: false
+    buffer_size: 4000 #4000 is minimum, and we are not using the on device speaker
+
 micro_wake_word:
   id: mww
   microphone:
@@ -410,6 +430,7 @@ micro_wake_word:
 voice_assistant:
   id: va
   microphone: mic
+  media_player: dummy_media_player
   micro_wake_word: mww
   use_wake_word: false
   noise_suppression_level: 3