Actually, revert the change making it so that STTService takes model and language args at init time. It'll be up to the subclasses to append those to _settings (or better yet, provide their own service-specific _settings). This avoids rocking the boat too too much.

.claude/skills/cleanup/SKILL.md

@@ -291,7 +291,7 @@ class NewTTSService(TTSService):
             voice: Voice identifier to use.
             **kwargs: Additional arguments passed to the parent service.
         """
-        super().__init__(voice=voice, **kwargs)
+        super().__init__(**kwargs)
         self._api_key = api_key
 ```
 

src/pipecat/services/asyncai/tts.py

@@ -153,7 +153,6 @@ class AsyncAITTSService(AudioContextTTSService):
             pause_frame_processing=True,
             push_stop_frames=True,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 
@@ -497,7 +496,7 @@ class AsyncAIHttpTTSService(TTSService):
             params: Additional input parameters for voice customization.
             **kwargs: Additional arguments passed to the parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or AsyncAIHttpTTSService.InputParams()
 

src/pipecat/services/aws/tts.py

@@ -195,7 +195,7 @@ class AWSPollyTTSService(TTSService):
             params: Additional input parameters for voice customization.
             **kwargs: Additional arguments passed to parent TTSService class.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or AWSPollyTTSService.InputParams()
 

src/pipecat/services/azure/tts.py

@@ -295,7 +295,6 @@ class AzureTTSService(WordTTSService, AzureBaseTTSService):
             push_stop_frames=True,
             pause_frame_processing=True,
             sample_rate=sample_rate,
-            voice=voice,
             **kwargs,
         )
 
@@ -734,7 +733,7 @@ class AzureHttpTTSService(TTSService, AzureBaseTTSService):
             params: Voice and synthesis parameters configuration.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         # Initialize Azure-specific functionality from mixin
         self._init_azure_base(api_key=api_key, region=region, voice=voice, params=params)

src/pipecat/services/camb/tts.py

@@ -213,7 +213,7 @@ class CambTTSService(TTSService):
             params: Additional voice parameters. If None, uses defaults.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         self._api_key = api_key
         self._timeout = timeout

src/pipecat/services/cartesia/tts.py

@@ -313,7 +313,6 @@ class CartesiaTTSService(AudioContextWordTTSService):
             pause_frame_processing=True,
             sample_rate=sample_rate,
             text_aggregator=text_aggregator,
-            voice=voice_id,
             **kwargs,
         )
 
@@ -721,7 +720,7 @@ class CartesiaHttpTTSService(TTSService):
             params: Additional input parameters for voice customization.
             **kwargs: Additional arguments passed to the parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or CartesiaHttpTTSService.InputParams()
 

src/pipecat/services/deepgram/tts.py

@@ -101,7 +101,6 @@ class DeepgramTTSService(WebsocketTTSService):
             pause_frame_processing=True,
             push_stop_frames=True,
             append_trailing_space=True,
-            voice=voice,
             **kwargs,
         )
 
@@ -388,7 +387,7 @@ class DeepgramHttpTTSService(TTSService):
             encoding: Audio encoding format. Defaults to "linear16".
             **kwargs: Additional arguments passed to parent TTSService class.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         self._api_key = api_key
         self._session = aiohttp_session

src/pipecat/services/elevenlabs/tts.py

@@ -400,7 +400,6 @@ class ElevenLabsTTSService(AudioContextWordTTSService):
             push_stop_frames=True,
             pause_frame_processing=True,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 
@@ -906,7 +905,6 @@ class ElevenLabsHttpTTSService(WordTTSService):
             push_text_frames=False,
             push_stop_frames=True,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 

src/pipecat/services/google/tts.py

@@ -602,7 +602,7 @@ class GoogleHttpTTSService(TTSService):
             params: Voice customization parameters including pitch, rate, volume, etc.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or GoogleHttpTTSService.InputParams()
 
@@ -1015,7 +1015,7 @@ class GoogleTTSService(GoogleBaseTTSService):
             params: Language configuration parameters.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or GoogleTTSService.InputParams()
 
@@ -1220,7 +1220,7 @@ class GeminiTTSService(GoogleBaseTTSService):
                 f"Google TTS only supports {self.GOOGLE_SAMPLE_RATE}Hz sample rate. "
                 f"Current rate of {sample_rate}Hz may cause issues."
             )
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or GeminiTTSService.InputParams()
 

src/pipecat/services/gradium/tts.py

@@ -91,7 +91,6 @@ class GradiumTTSService(InterruptibleWordTTSService):
             push_stop_frames=True,
             pause_frame_processing=True,
             sample_rate=SAMPLE_RATE,
-            voice=voice_id,
             **kwargs,
         )
 

src/pipecat/services/groq/tts.py

@@ -102,7 +102,6 @@ class GroqTTSService(TTSService):
         super().__init__(
             pause_frame_processing=True,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 

src/pipecat/services/hathora/tts.py

@@ -109,7 +109,6 @@ class HathoraTTSService(TTSService):
         """
         super().__init__(
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
         self._model = model

src/pipecat/services/hume/tts.py

@@ -126,7 +126,6 @@ class HumeTTSService(WordTTSService):
             sample_rate=sample_rate,
             push_text_frames=False,
             push_stop_frames=True,
-            voice=voice_id,
             **kwargs,
         )
 

src/pipecat/services/inworld/tts.py

@@ -144,7 +144,6 @@ class InworldHttpTTSService(WordTTSService):
             push_text_frames=False,
             push_stop_frames=True,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 
@@ -519,7 +518,6 @@ class InworldTTSService(AudioContextWordTTSService):
             sample_rate=sample_rate,
             aggregate_sentences=aggregate_sentences,
             append_trailing_space=append_trailing_space,
-            voice=voice_id,
             **kwargs,
         )
 

src/pipecat/services/kokoro/tts.py

@@ -137,7 +137,7 @@ class KokoroTTSService(TTSService):
             **kwargs: Additional arguments passed to parent `TTSService`.
 
         """
-        super().__init__(voice=voice_id, **kwargs)
+        super().__init__(**kwargs)
 
         params = params or KokoroTTSService.InputParams()
 

src/pipecat/services/lmnt/tts.py

@@ -118,7 +118,6 @@ class LmntTTSService(InterruptibleTTSService):
             push_stop_frames=True,
             pause_frame_processing=True,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 

src/pipecat/services/minimax/tts.py

@@ -227,7 +227,7 @@ class MiniMaxHttpTTSService(TTSService):
             params: Additional configuration parameters.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or MiniMaxHttpTTSService.InputParams()
 

src/pipecat/services/neuphonic/tts.py

@@ -139,7 +139,6 @@ class NeuphonicTTSService(InterruptibleTTSService):
             push_stop_frames=True,
             stop_frame_timeout_s=2.0,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 
@@ -443,7 +442,7 @@ class NeuphonicHttpTTSService(TTSService):
             params: Additional input parameters for TTS configuration.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or NeuphonicHttpTTSService.InputParams()
 

src/pipecat/services/openai/tts.py

@@ -132,7 +132,7 @@ class OpenAITTSService(TTSService):
                 f"OpenAI TTS only supports {self.OPENAI_SAMPLE_RATE}Hz sample rate. "
                 f"Current rate of {sample_rate}Hz may cause issues."
             )
-        super().__init__(sample_rate=sample_rate, voice=voice, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         self.set_model_name(model)
         self._client = AsyncOpenAI(api_key=api_key, base_url=base_url)

src/pipecat/services/playht/tts.py

@@ -173,7 +173,6 @@ class PlayHTTTSService(InterruptibleTTSService):
         super().__init__(
             pause_frame_processing=True,
             sample_rate=sample_rate,
-            voice=voice_url,
             **kwargs,
         )
 
@@ -511,7 +510,7 @@ class PlayHTHttpTTSService(TTSService):
             params: Additional input parameters for voice customization.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_url, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         # Warn about deprecated protocol parameter if explicitly provided
         if protocol:

src/pipecat/services/resembleai/tts.py

@@ -94,7 +94,6 @@ class ResembleAITTSService(AudioContextWordTTSService):
         """
         super().__init__(
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 

src/pipecat/services/rime/tts.py

@@ -191,7 +191,6 @@ class RimeTTSService(AudioContextWordTTSService):
             pause_frame_processing=True,
             append_trailing_space=True,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
 
@@ -582,7 +581,7 @@ class RimeHttpTTSService(TTSService):
             params: Additional configuration parameters.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         params = params or RimeHttpTTSService.InputParams()
 
@@ -762,7 +761,6 @@ class RimeNonJsonTTSService(InterruptibleTTSService):
             aggregate_sentences=aggregate_sentences,
             push_stop_frames=True,
             pause_frame_processing=True,
-            voice=voice_id,
             **kwargs,
         )
         params = params or RimeNonJsonTTSService.InputParams()

src/pipecat/services/sarvam/tts.py

@@ -464,7 +464,7 @@ class SarvamHttpTTSService(TTSService):
         if voice_id is None:
             voice_id = self._config.default_speaker
 
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         self._api_key = api_key
         self._base_url = base_url
@@ -821,7 +821,6 @@ class SarvamTTSService(InterruptibleTTSService):
             pause_frame_processing=True,
             push_stop_frames=True,
             sample_rate=sample_rate,
-            voice=voice_id,
             **kwargs,
         )
         params = params or SarvamTTSService.InputParams()

src/pipecat/services/tts_service.py

@@ -145,11 +145,6 @@ class TTSService(AIService):
         text_filter: Optional[BaseTextFilter] = None,
         # Audio transport destination of the generated frames.
         transport_destination: Optional[str] = None,
-        # Voice identifier or name to use for speech synthesis
-        voice: Optional[str] = None,
-        # Language to use for speech synthesis. This will be translated to a
-        # service-specific language identifier before being applied
-        language: Optional[Language] = None,
         **kwargs,
     ):
         """Initialize the TTS service.
@@ -183,10 +178,6 @@ class TTSService(AIService):
                     Use `text_filters` instead, which allows multiple filters.
 
             transport_destination: Destination for generated audio frames.
-            voice: Voice identifier or name to use for speech synthesis.
-            language: Language to use for speech synthesis. This will be
-                translated to a service-specific language identifier before
-                being applied.
             **kwargs: Additional arguments passed to the parent AIService.
         """
         super().__init__(**kwargs)
@@ -201,7 +192,7 @@ class TTSService(AIService):
         self._init_sample_rate = sample_rate
         self._sample_rate = 0
         self._settings = TTSSettings(
-            voice=voice, language=language
+            voice=""
         )  # Here in case subclass doesn't implement more specific settings (hopefully shouldn't happen)
         self._text_aggregator: BaseTextAggregator = text_aggregator or SimpleTextAggregator()
         if text_aggregator:

src/pipecat/services/xtts/tts.py

@@ -111,7 +111,7 @@ class XTTSService(TTSService):
             sample_rate: Audio sample rate. If None, uses default.
             **kwargs: Additional arguments passed to parent TTSService.
         """
-        super().__init__(sample_rate=sample_rate, voice=voice_id, **kwargs)
+        super().__init__(sample_rate=sample_rate, **kwargs)
 
         self._settings = XTTSTTSSettings(
             voice=voice_id,