Allow TTS requests to resolve in the background (#90944)

homeassistant/components/cloud/tts.py

@@ -4,11 +4,16 @@ from hass_nabucasa import Cloud
 from hass_nabucasa.voice import MAP_VOICE, AudioOutput, VoiceError
 import voluptuous as vol
 
-from homeassistant.components.tts import CONF_LANG, PLATFORM_SCHEMA, Provider
+from homeassistant.components.tts import (
+    ATTR_AUDIO_OUTPUT,
+    CONF_LANG,
+    PLATFORM_SCHEMA,
+    Provider,
+)
 
 from .const import DOMAIN
 
-CONF_GENDER = "gender"
+ATTR_GENDER = "gender"
 
 SUPPORT_LANGUAGES = list({key[0] for key in MAP_VOICE})
 
@@ -18,8 +23,8 @@ def validate_lang(value):
     if (lang := value.get(CONF_LANG)) is None:
         return value
 
-    if (gender := value.get(CONF_GENDER)) is None:
+    if (gender := value.get(ATTR_GENDER)) is None:
-        gender = value[CONF_GENDER] = next(
+        gender = value[ATTR_GENDER] = next(
             (chk_gender for chk_lang, chk_gender in MAP_VOICE if chk_lang == lang), None
         )
 
@@ -33,7 +38,7 @@ PLATFORM_SCHEMA = vol.All(
     PLATFORM_SCHEMA.extend(
         {
             vol.Optional(CONF_LANG): str,
-            vol.Optional(CONF_GENDER): str,
+            vol.Optional(ATTR_GENDER): str,
         }
     ),
     validate_lang,
@@ -49,7 +54,7 @@ async def async_get_engine(hass, config, discovery_info=None):
         gender = None
     else:
         language = config[CONF_LANG]
-        gender = config[CONF_GENDER]
+        gender = config[ATTR_GENDER]
 
     return CloudProvider(cloud, language, gender)
 
@@ -87,12 +92,15 @@ class CloudProvider(Provider):
     @property
     def supported_options(self):
         """Return list of supported options like voice, emotion."""
-        return [CONF_GENDER]
+        return [ATTR_GENDER, ATTR_AUDIO_OUTPUT]
 
     @property
     def default_options(self):
         """Return a dict include default options."""
-        return {CONF_GENDER: self._gender}
+        return {
+            ATTR_GENDER: self._gender,
+            ATTR_AUDIO_OUTPUT: AudioOutput.MP3,
+        }
 
     async def async_get_tts_audio(self, message, language, options=None):
         """Load TTS from NabuCasa Cloud."""
@@ -101,10 +109,10 @@ class CloudProvider(Provider):
             data = await self.cloud.voice.process_tts(
                 message,
                 language,
-                gender=options[CONF_GENDER],
+                gender=options[ATTR_GENDER],
-                output=AudioOutput.MP3,
+                output=options[ATTR_AUDIO_OUTPUT],
             )
         except VoiceError:
             return (None, None)
 
-        return ("mp3", data)
+        return (str(options[ATTR_AUDIO_OUTPUT]), data)

homeassistant/components/tts/__init__.py

@@ -59,6 +59,7 @@ ATTR_LANGUAGE = "language"
 ATTR_MESSAGE = "message"
 ATTR_OPTIONS = "options"
 ATTR_PLATFORM = "platform"
+ATTR_AUDIO_OUTPUT = "audio_output"
 
 BASE_URL_KEY = "tts_base_url"
 
@@ -134,6 +135,7 @@ class TTSCache(TypedDict):
 
     filename: str
     voice: bytes
+    pending: asyncio.Task | None
 
 
 @callback
@@ -495,8 +497,11 @@ class SpeechManager:
                 )
 
         extension = os.path.splitext(self.mem_cache[cache_key]["filename"])[1][1:]
-        data = self.mem_cache[cache_key]["voice"]
+        cached = self.mem_cache[cache_key]
-        return extension, data
+        if pending := cached.get("pending"):
+            await pending
+            cached = self.mem_cache[cache_key]
+        return extension, cached["voice"]
 
     @callback
     def _generate_cache_key(
@@ -527,7 +532,17 @@ class SpeechManager:
         This method is a coroutine.
         """
         provider = self.providers[engine]
-        extension, data = await provider.async_get_tts_audio(message, language, options)
+
+        if options is not None and ATTR_AUDIO_OUTPUT in options:
+            expected_extension = options[ATTR_AUDIO_OUTPUT]
+        else:
+            expected_extension = None
+
+        async def get_tts_data() -> str:
+            """Handle data available."""
+            extension, data = await provider.async_get_tts_audio(
+                message, language, options
+            )
 
             if data is None or extension is None:
                 raise HomeAssistantError(f"No TTS from {engine} for '{message}'")
@@ -543,7 +558,9 @@ class SpeechManager:
 
             # Save to memory
             if extension == "mp3":
-            data = self.write_tags(filename, data, provider, message, language, options)
+                data = self.write_tags(
+                    filename, data, provider, message, language, options
+                )
             self._async_store_to_memcache(cache_key, filename, data)
 
             if cache:
@@ -553,6 +570,26 @@ class SpeechManager:
 
             return filename
 
+        audio_task = self.hass.async_create_task(get_tts_data())
+
+        if expected_extension is None:
+            return await audio_task
+
+        def handle_error(_future: asyncio.Future) -> None:
+            """Handle error."""
+            if audio_task.exception():
+                self.mem_cache.pop(cache_key, None)
+
+        audio_task.add_done_callback(handle_error)
+
+        filename = f"{cache_key}.{expected_extension}".lower()
+        self.mem_cache[cache_key] = {
+            "filename": filename,
+            "voice": b"",
+            "pending": audio_task,
+        }
+        return filename
+
     async def _async_save_tts_audio(
         self, cache_key: str, filename: str, data: bytes
     ) -> None:
@@ -601,7 +638,11 @@ class SpeechManager:
         self, cache_key: str, filename: str, data: bytes
     ) -> None:
         """Store data to memcache and set timer to remove it."""
-        self.mem_cache[cache_key] = {"filename": filename, "voice": data}
+        self.mem_cache[cache_key] = {
+            "filename": filename,
+            "voice": data,
+            "pending": None,
+        }
 
         @callback
         def async_remove_from_mem() -> None:
@@ -628,7 +669,11 @@ class SpeechManager:
             await self._async_file_to_mem(cache_key)
 
         content, _ = mimetypes.guess_type(filename)
-        return content, self.mem_cache[cache_key]["voice"]
+        cached = self.mem_cache[cache_key]
+        if pending := cached.get("pending"):
+            await pending
+            cached = self.mem_cache[cache_key]
+        return content, cached["voice"]
 
     @staticmethod
     def write_tags(

tests/components/cloud/test_tts.py

@@ -58,17 +58,17 @@ async def test_prefs_default_voice(
     )
 
     assert provider_pref.default_language == "en-US"
-    assert provider_pref.default_options == {"gender": "female"}
+    assert provider_pref.default_options == {"gender": "female", "audio_output": "mp3"}
     assert provider_conf.default_language == "fr-FR"
-    assert provider_conf.default_options == {"gender": "female"}
+    assert provider_conf.default_options == {"gender": "female", "audio_output": "mp3"}
 
     await cloud_prefs.async_update(tts_default_voice=("nl-NL", "male"))
     await hass.async_block_till_done()
 
     assert provider_pref.default_language == "nl-NL"
-    assert provider_pref.default_options == {"gender": "male"}
+    assert provider_pref.default_options == {"gender": "male", "audio_output": "mp3"}
     assert provider_conf.default_language == "fr-FR"
-    assert provider_conf.default_options == {"gender": "female"}
+    assert provider_conf.default_options == {"gender": "female", "audio_output": "mp3"}
 
 
 async def test_provider_properties(cloud_with_prefs) -> None:
@@ -76,7 +76,7 @@ async def test_provider_properties(cloud_with_prefs) -> None:
     provider = await tts.async_get_engine(
         Mock(data={const.DOMAIN: cloud_with_prefs}), None, {}
     )
-    assert provider.supported_options == ["gender"]
+    assert provider.supported_options == ["gender", "audio_output"]
     assert "nl-NL" in provider.supported_languages
 
 
@@ -85,5 +85,5 @@ async def test_get_tts_audio(cloud_with_prefs) -> None:
     provider = await tts.async_get_engine(
         Mock(data={const.DOMAIN: cloud_with_prefs}), None, {}
     )
-    assert provider.supported_options == ["gender"]
+    assert provider.supported_options == ["gender", "audio_output"]
     assert "nl-NL" in provider.supported_languages

tests/components/tts/test_init.py

@@ -1,4 +1,5 @@
 """The tests for the TTS component."""
+import asyncio
 from http import HTTPStatus
 from typing import Any
 from unittest.mock import patch
@@ -996,3 +997,73 @@ async def test_support_options(hass: HomeAssistant, setup_tts) -> None:
         await tts.async_support_options(hass, "test", "en", {"invalid_option": "yo"})
         is False
     )
+
+
+async def test_fetching_in_async(hass: HomeAssistant, hass_client) -> None:
+    """Test async fetching of data."""
+    tts_audio = asyncio.Future()
+
+    class ProviderWithAsyncFetching(MockProvider):
+        """Provider that supports audio output option."""
+
+        @property
+        def supported_options(self) -> list[str]:
+            """Return list of supported options like voice, emotions."""
+            return [tts.ATTR_AUDIO_OUTPUT]
+
+        @property
+        def default_options(self) -> dict[str, str]:
+            """Return a dict including the default options."""
+            return {tts.ATTR_AUDIO_OUTPUT: "mp3"}
+
+        async def async_get_tts_audio(
+            self, message: str, language: str, options: dict[str, Any] | None = None
+        ) -> tts.TtsAudioType:
+            return ("mp3", await tts_audio)
+
+    mock_integration(hass, MockModule(domain="test"))
+    mock_platform(hass, "test.tts", MockTTS(ProviderWithAsyncFetching))
+    assert await async_setup_component(hass, tts.DOMAIN, {"tts": {"platform": "test"}})
+
+    # Test async_get_media_source_audio
+    media_source_id = tts.generate_media_source_id(
+        hass, "test message", "test", "en", None, None
+    )
+
+    task = hass.async_create_task(
+        tts.async_get_media_source_audio(hass, media_source_id)
+    )
+    task2 = hass.async_create_task(
+        tts.async_get_media_source_audio(hass, media_source_id)
+    )
+
+    url = await get_media_source_url(hass, media_source_id)
+    client = await hass_client()
+    client_get_task = hass.async_create_task(client.get(url))
+
+    # Make sure that tasks are waiting for our future to resolve
+    done, pending = await asyncio.wait((task, task2, client_get_task), timeout=0.1)
+    assert len(done) == 0
+    assert len(pending) == 3
+
+    tts_audio.set_result(b"test")
+
+    assert await task == ("mp3", b"test")
+    assert await task2 == ("mp3", b"test")
+
+    req = await client_get_task
+    assert req.status == HTTPStatus.OK
+    assert await req.read() == b"test"
+
+    # Test error is not cached
+    media_source_id = tts.generate_media_source_id(
+        hass, "test message 2", "test", "en", None, None
+    )
+    tts_audio = asyncio.Future()
+    tts_audio.set_exception(HomeAssistantError("test error"))
+    with pytest.raises(HomeAssistantError):
+        assert await tts.async_get_media_source_audio(hass, media_source_id)
+
+    tts_audio = asyncio.Future()
+    tts_audio.set_result(b"test 2")
+    await tts.async_get_media_source_audio(hass, media_source_id) == ("mp3", b"test 2")