diff --git a/homeassistant/components/tts/__init__.py b/homeassistant/components/tts/__init__.py
index 3645afedd6d..fb9dfcac13c 100644
--- a/homeassistant/components/tts/__init__.py
+++ b/homeassistant/components/tts/__init__.py
@@ -527,6 +527,8 @@ class ResultStream:
 
         This method will leverage a disk cache to speed up generation.
         """
+        if self._result_cache.done():
+            return
         self._result_cache.set_result(
             self._manager.async_cache_message_in_memory(
                 engine=self.engine,
@@ -543,6 +545,8 @@ class ResultStream:
 
         This method can result in faster first byte when generating long responses.
         """
+        if self._result_cache.done():
+            return
         self._result_cache.set_result(
             self._manager.async_cache_message_stream_in_memory(
                 engine=self.engine,
diff --git a/tests/components/tts/test_init.py b/tests/components/tts/test_init.py
index 5a6e988c82d..ee7878e603a 100644
--- a/tests/components/tts/test_init.py
+++ b/tests/components/tts/test_init.py
@@ -1954,6 +1954,46 @@ async def test_stream(hass: HomeAssistant, mock_tts_entity: MockTTSEntity) -> No
     assert result_data == data
 
 
+async def test_result_stream_message_set_idempotent(
+    hass: HomeAssistant, mock_tts_entity: MockTTSEntity
+) -> None:
+    """Test setting a result stream message more than once."""
+    await mock_config_entry_setup(hass, mock_tts_entity)
+
+    stream = tts.async_create_stream(hass, mock_tts_entity.entity_id)
+    stream.async_set_message("hello")
+    cache_first = stream._result_cache.result()
+    stream.async_set_message("world")
+    assert stream._result_cache.result() is cache_first
+
+    async def async_stream_tts_audio(
+        request: tts.TTSAudioRequest,
+    ) -> tts.TTSAudioResponse:
+        """Mock stream TTS audio."""
+
+        async def gen_data():
+            async for msg in request.message_gen:
+                yield msg.encode()
+
+        return tts.TTSAudioResponse(
+            extension="mp3",
+            data_gen=gen_data(),
+        )
+
+    mock_tts_entity.async_stream_tts_audio = async_stream_tts_audio
+    mock_tts_entity.async_supports_streaming_input = Mock(return_value=True)
+
+    async def stream_message():
+        """Mock stream message."""
+        yield "h"
+
+    stream2 = tts.async_create_stream(hass, mock_tts_entity.entity_id)
+    stream2.async_set_message_stream(stream_message())
+    cache_first = stream2._result_cache.result()
+    stream2.async_set_message_stream(stream_message())
+    assert stream2._result_cache.result() is cache_first
+
+
 async def test_tts_cache() -> None:
     """Test TTSCache."""