Merge pull request #18 from raldone01/feat/support_custom_endpoints

Support custom openai compatible endpoints

custom_components/openai_tts/config_flow.py

@@ -3,29 +3,24 @@
 from typing import Any
 import voluptuous as vol
 import logging
+from urllib.parse import urlparse
 
 from homeassistant import data_entry_flow
 from homeassistant.config_entries import ConfigFlow
 from homeassistant.helpers.selector import selector
 from homeassistant.exceptions import HomeAssistantError
 
-from .const import CONF_API_KEY, CONF_MODEL, CONF_VOICE, CONF_SPEED, DOMAIN, MODELS, VOICES
+from .const import CONF_API_KEY, CONF_MODEL, CONF_VOICE, CONF_SPEED, CONF_URL, DOMAIN, MODELS, VOICES, UNIQUE_ID
 
 _LOGGER = logging.getLogger(__name__)
 
-class WrongAPIKey(HomeAssistantError):
-    """Error to indicate no or wrong API key."""
-
-async def validate_api_key(api_key: str):
-    """Validate the API key format."""
-    if api_key is None:
-        raise WrongAPIKey("API key is required")
-    if not (51 <= len(api_key) <= 70):
-        raise WrongAPIKey("Invalid API key length")
+def generate_unique_id(user_input: dict) -> str:
+    """Generate a unique id from user input."""
+    url = urlparse(user_input[CONF_URL])
+    return f"{url.hostname}_{user_input[CONF_MODEL]}_{user_input[CONF_VOICE]}"
 
 async def validate_user_input(user_input: dict):
     """Validate user input fields."""
-    await validate_api_key(user_input.get(CONF_API_KEY))
     if user_input.get(CONF_MODEL) is None:
         raise ValueError("Model is required")
     if user_input.get(CONF_VOICE) is None:
@@ -35,22 +30,23 @@ class OpenAITTSConfigFlow(ConfigFlow, domain=DOMAIN):
     """Handle a config flow for OpenAI TTS."""
     VERSION = 1
     data_schema = vol.Schema({
-        vol.Required(CONF_API_KEY): str,
+        vol.Optional(CONF_API_KEY): str,
+        vol.Optional(CONF_URL, default="https://api.openai.com/v1/audio/speech"): str,
         vol.Optional(CONF_SPEED, default=1.0): vol.Coerce(float),
         vol.Required(CONF_MODEL, default="tts-1"): selector({
             "select": {
                 "options": MODELS,
                 "mode": "dropdown",
                 "sort": True,
-                "custom_value": False
+                "custom_value": True
             }
         }),
         vol.Required(CONF_VOICE, default="shimmer"): selector({
             "select": {
                 "options": VOICES,
                 "mode": "dropdown",
                 "sort": True,
-                "custom_value": False
+                "custom_value": True
             }
         })
     })
@@ -61,14 +57,17 @@ async def async_step_user(self, user_input: dict[str, Any] | None = None):
         if user_input is not None:
             try:
                 await validate_user_input(user_input)
-                await self.async_set_unique_id(f"{user_input[CONF_VOICE]}_{user_input[CONF_MODEL]}")
+                unique_id = generate_unique_id(user_input)
+                user_input[UNIQUE_ID] = unique_id
+                await self.async_set_unique_id(unique_id)
                 self._abort_if_unique_id_configured()
-                return self.async_create_entry(title="OpenAI TTS", data=user_input)
+                hostname = urlparse(user_input[CONF_URL]).hostname
+                return self.async_create_entry(title=f"OpenAI TTS ({hostname}, {user_input[CONF_MODEL]}, {user_input[CONF_VOICE]})", data=user_input)
             except data_entry_flow.AbortFlow:
                 return self.async_abort(reason="already_configured")
             except HomeAssistantError as e:
                 _LOGGER.exception(str(e))
-                errors["api_key"] = "wrong_api_key"
+                errors["base"] = str(e)
             except ValueError as e:
                 _LOGGER.exception(str(e))
                 errors["base"] = str(e)

custom_components/openai_tts/const.py

@@ -5,6 +5,7 @@
 CONF_MODEL = 'model'
 CONF_VOICE = 'voice'
 CONF_SPEED = 'speed'
+CONF_URL = 'url'
+UNIQUE_ID = 'unique_id'
 MODELS = ["tts-1", "tts-1-hd"]
 VOICES = ["alloy", "echo", "fable", "onyx", "nova", "shimmer"]
-URL = "https://api.openai.com/v1/audio/speech"

custom_components/openai_tts/manifest.json

@@ -1,12 +1,16 @@
 {
   "domain": "openai_tts",
   "name": "OpenAI TTS",
-  "codeowners": ["@sfortis"],
+  "codeowners": [
+    "@sfortis"
+  ],
   "config_flow": true,
   "dependencies": [],
   "documentation": "https://github.com/sfortis/openai_tts/",
   "iot_class": "cloud_polling",
   "issue_tracker": "https://github.com/sfortis/openai_tts/issues",
-  "requirements": ["requests>=2.25.1"],
-  "version": "0.2.1"
+  "requirements": [
+    "requests>=2.25.1"
+  ],
+  "version": "0.2.2"
 }

custom_components/openai_tts/openaitts_engine.py

@@ -1,26 +1,27 @@
 import requests
 
-from .const import URL
-
-
 class OpenAITTSEngine:
 
-    def __init__(self, api_key: str, voice: str, model: str, speed: int):
+    def __init__(self, api_key: str, voice: str, model: str, speed: int, url: str):
         self._api_key = api_key
         self._voice = voice
         self._model = model
         self._speed = speed
-        self._url = URL
+        self._url = url
 
     def get_tts(self, text: str):
         """ Makes request to OpenAI TTS engine to convert text into audio"""
-        headers: dict = {"Authorization": f"Bearer {self._api_key}"}
-        data: dict = {"model": self._model, "input": text, "voice": self._voice, "speed": self._speed}
+        headers: dict = {"Authorization": f"Bearer {self._api_key}"} if self._api_key else {}
+        data: dict = {
+            "model": self._model,
+            "input": text,
+            "voice": self._voice,
+            "response_format": "wav",
+            "speed": self._speed
+        }
         return requests.post(self._url, headers=headers, json=data)
 
     @staticmethod
     def get_supported_langs() -> list:
         """Returns list of supported languages. Note: the model determines the provides language automatically."""
         return ["af", "ar", "hy", "az", "be", "bs", "bg", "ca", "zh", "hr", "cs", "da", "nl", "en", "et", "fi", "fr", "gl", "de", "el", "he", "hi", "hu", "is", "id", "it", "ja", "kn", "kk", "ko", "lv", "lt", "mk", "ms", "mr", "mi", "ne", "no", "fa", "pl", "pt", "ro", "ru", "sr", "sk", "sl", "es", "sw", "sv", "tl", "ta", "th", "tr", "uk", "ur", "vi", "cy"]
-
-

custom_components/openai_tts/strings.json

@@ -8,16 +8,17 @@
           "api_key": "Enter OpenAI API key.",
           "speed": "Enter speed of the speech",
           "model": "Select model to be used.",
-          "voice": "Select voice."
+          "voice": "Select voice.",
+          "url": "Enter the OpenAI-compatible endpoint. Optionally include a port number."
         }
       }
     },
     "error": {
       "wrong_api_key": "Invalid API key. Please enter a valid API key.",
-      "already_configured": "This voice is already configured."
+      "already_configured": "This voice and endpoint are already configured."
     },
     "abort": {
-      "already_configured": "This voice is already configured."
+      "already_configured": "This voice and endpoint are already configured."
     }
   }
 }

custom_components/openai_tts/translations/cs.json

@@ -8,7 +8,8 @@
           "api_key": "Vlož OpenAI API klíč.",
           "speed": "Vlož rychlost řeči.",
           "model": "Vyber model k použití.",
-          "voice": "Vyber hlas."
+          "voice": "Vyber hlas.",
+          "url": "Zadejte koncový bod kompatibilní s OpenAI. Volitelně uveďte číslo portu."
         }
       }
     },

custom_components/openai_tts/translations/de.json

@@ -0,0 +1,24 @@
+{
+  "config": {
+    "step": {
+      "user": {
+        "title": "Füge eine Text zu Sprache Engine hinzu",
+        "description": "Gib Konfigurationsdaten ein. Schau in die Dokumentation für weitere Informationen.",
+        "data": {
+          "api_key": "Gib den OpenAI API Schlüssel ein.",
+          "speed": "Gib die Geschwindigkeit der Sprache ein",
+          "model": "Wähle das zu verwendende Modell.",
+          "voice": "Wähle eine Stimme.",
+          "url": "Gib den OpenAI-kompatiblen Endpunkt ein. Optional kann eine Portnummer angegeben werden."
+        }
+      }
+    },
+    "error": {
+      "wrong_api_key": "Ungültiger API Schlüssel. Bitte gib einen gültigen API Schlüssel ein.",
+      "already_configured": "Diese Stimme und Endpunkt sind bereits konfiguriert."
+    },
+    "abort": {
+      "already_configured": "Diese Stimme und Endpunkt sind bereits konfiguriert."
+    }
+  }
+}

custom_components/openai_tts/translations/en.json

@@ -8,16 +8,17 @@
           "api_key": "Enter OpenAI API key.",
           "speed": "Enter speed of the speech",
           "model": "Select model to be used.",
-          "voice": "Select voice."
+          "voice": "Select voice.",
+          "url": "Enter the OpenAI-compatible endpoint. Optionally include a port number."
         }
       }
     },
     "error": {
       "wrong_api_key": "Invalid API key. Please enter a valid API key.",
-      "already_configured": "This voice is already configured."
+      "already_configured": "This voice and endpoint are already configured."
     },
     "abort": {
-      "already_configured": "This voice is already configured."
+      "already_configured": "This voice and endpoint are already configured."
     }
   }
 }

custom_components/openai_tts/tts.py

@@ -7,24 +7,29 @@
 from homeassistant.core import HomeAssistant
 from homeassistant.helpers.entity_platform import AddEntitiesCallback
 from homeassistant.helpers.entity import generate_entity_id
-from .const import CONF_API_KEY, CONF_MODEL, CONF_SPEED, CONF_VOICE, DOMAIN
+from .const import CONF_API_KEY, CONF_MODEL, CONF_SPEED, CONF_VOICE, CONF_URL, DOMAIN, UNIQUE_ID
 from .openaitts_engine import OpenAITTSEngine
 from homeassistant.exceptions import MaxLengthExceeded
 
 _LOGGER = logging.getLogger(__name__)
 
-
 async def async_setup_entry(
     hass: HomeAssistant,
     config_entry: ConfigEntry,
     async_add_entities: AddEntitiesCallback,
 ) -> None:
     """Set up OpenAI Text-to-speech platform via config entry."""
+
+    api_key = None
+    if CONF_API_KEY in config_entry.data:
+        api_key = config_entry.data[CONF_API_KEY]
+
     engine = OpenAITTSEngine(
-        config_entry.data[CONF_API_KEY],
+        api_key,
         config_entry.data[CONF_VOICE],
         config_entry.data[CONF_MODEL],
         config_entry.data[CONF_SPEED],
+        config_entry.data[CONF_URL]
     )
     async_add_entities([OpenAITTSEntity(hass, config_entry, engine)])
 
@@ -39,7 +44,11 @@ def __init__(self, hass, config, engine):
         self.hass = hass
         self._engine = engine
         self._config = config
-        self._attr_unique_id = f"{config.data[CONF_VOICE]}_{config.data[CONF_MODEL]}"
+
+        self._attr_unique_id = config.data.get(UNIQUE_ID)
+        if self._attr_unique_id is None:
+            # generate a legacy unique_id
+            self._attr_unique_id = f"{config.data[CONF_VOICE]}_{config.data[CONF_MODEL]}"
         self.entity_id = generate_entity_id("tts.openai_tts_{}", config.data[CONF_VOICE], hass=hass)
 
     @property
@@ -74,7 +83,7 @@ def get_tts_audio(self, message, language, options=None):
             speech = self._engine.get_tts(message)
 
             # The response should contain the audio file content
-            return "mp3", speech.content
+            return "wav", speech.content
         except MaxLengthExceeded:
             _LOGGER.error("Maximum length of the message exceeded")
         except Exception as e:

hacs.json

@@ -1,5 +1,6 @@
 {
     "name": "OpenAI TTS Speech Service",
     "homeassistant": "2024.5.3",
-    "render_readme": true
+    "render_readme": true,
+    "version": "0.2.3"
 }