diff --git a/src/pipecat/services/asyncai/tts.py b/src/pipecat/services/asyncai/tts.py
index 42c0a09a4..d2ac74445 100644
--- a/src/pipecat/services/asyncai/tts.py
+++ b/src/pipecat/services/asyncai/tts.py
@@ -171,9 +171,7 @@ class AsyncAITTSService(WebsocketTTSService):
         if params is not None:
             self._warn_init_param_moved_to_settings("params")
             if not settings:
-                default_settings.language = (
-                    self.language_to_service_language(params.language) if params.language else None
-                )
+                default_settings.language = params.language
 
         # 4. Apply settings delta (canonical API, always wins)
         if settings is not None:
@@ -565,9 +563,7 @@ class AsyncAIHttpTTSService(TTSService):
         if params is not None:
             self._warn_init_param_moved_to_settings("params")
             if not settings:
-                default_settings.language = (
-                    self.language_to_service_language(params.language) if params.language else None
-                )
+                default_settings.language = params.language
 
         # 4. Apply settings delta (canonical API, always wins)
         if settings is not None:
diff --git a/src/pipecat/services/aws/stt.py b/src/pipecat/services/aws/stt.py
index eed1a321d..ace05090d 100644
--- a/src/pipecat/services/aws/stt.py
+++ b/src/pipecat/services/aws/stt.py
@@ -100,13 +100,13 @@ class AWSTranscribeSTTService(WebsocketSTTService):
         # 1. Initialize default_settings with hardcoded defaults
         default_settings = self.Settings(
             model=None,
-            language=self.language_to_service_language(Language.EN),
+            language=Language.EN,
         )
 
         # 2. Apply direct init arg overrides (deprecated)
         if language is not None:
             self._warn_init_param_moved_to_settings("language", "language")
-            default_settings.language = self.language_to_service_language(language)
+            default_settings.language = language
 
         # 3. (No step 3, as there's no params object to apply)
 
diff --git a/src/pipecat/services/aws/tts.py b/src/pipecat/services/aws/tts.py
index 9648c45c8..32266886e 100644
--- a/src/pipecat/services/aws/tts.py
+++ b/src/pipecat/services/aws/tts.py
@@ -230,11 +230,7 @@ class AWSPollyTTSService(TTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 default_settings.engine = params.engine
-                default_settings.language = (
-                    self.language_to_service_language(params.language)
-                    if params.language
-                    else "en-US"
-                )
+                default_settings.language = params.language if params.language else "en-US"
                 default_settings.pitch = params.pitch
                 default_settings.rate = params.rate
                 default_settings.volume = params.volume
diff --git a/src/pipecat/services/azure/stt.py b/src/pipecat/services/azure/stt.py
index 857b166cb..57306e06a 100644
--- a/src/pipecat/services/azure/stt.py
+++ b/src/pipecat/services/azure/stt.py
@@ -106,13 +106,13 @@ class AzureSTTService(STTService):
         # 1. Initialize default_settings with hardcoded defaults
         default_settings = self.Settings(
             model=None,
-            language=language_to_azure_language(Language.EN_US),
+            language=Language.EN_US,
         )
 
         # 2. Apply direct init arg overrides (deprecated)
         if language is not None and language != Language.EN_US:
             self._warn_init_param_moved_to_settings("language", "language")
-            default_settings.language = language_to_azure_language(language)
+            default_settings.language = language
 
         # 3. (No step 3, as there's no params object to apply)
 
diff --git a/src/pipecat/services/azure/tts.py b/src/pipecat/services/azure/tts.py
index 0130ef5cb..a9491e9aa 100644
--- a/src/pipecat/services/azure/tts.py
+++ b/src/pipecat/services/azure/tts.py
@@ -312,11 +312,7 @@ class AzureTTSService(TTSService, AzureBaseTTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 default_settings.emphasis = params.emphasis
-                default_settings.language = (
-                    self.language_to_service_language(params.language)
-                    if params.language
-                    else "en-US"
-                )
+                default_settings.language = params.language if params.language else "en-US"
                 default_settings.pitch = params.pitch
                 default_settings.rate = params.rate
                 default_settings.role = params.role
@@ -809,11 +805,7 @@ class AzureHttpTTSService(TTSService, AzureBaseTTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 default_settings.emphasis = params.emphasis
-                default_settings.language = (
-                    self.language_to_service_language(params.language)
-                    if params.language
-                    else "en-US"
-                )
+                default_settings.language = params.language if params.language else "en-US"
                 default_settings.pitch = params.pitch
                 default_settings.rate = params.rate
                 default_settings.role = params.role
diff --git a/src/pipecat/services/camb/tts.py b/src/pipecat/services/camb/tts.py
index 9c360ed37..b6b83a928 100644
--- a/src/pipecat/services/camb/tts.py
+++ b/src/pipecat/services/camb/tts.py
@@ -260,9 +260,7 @@ class CambTTSService(TTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = (
-                        self.language_to_service_language(params.language) or "en-us"
-                    )
+                    default_settings.language = params.language
                 if params.user_instructions is not None:
                     default_settings.user_instructions = params.user_instructions
 
diff --git a/src/pipecat/services/cartesia/tts.py b/src/pipecat/services/cartesia/tts.py
index aca5c46c6..b713a0d9a 100644
--- a/src/pipecat/services/cartesia/tts.py
+++ b/src/pipecat/services/cartesia/tts.py
@@ -302,7 +302,7 @@ class CartesiaTTSService(WebsocketTTSService):
         default_settings = self.Settings(
             model="sonic-3",
             voice=None,
-            language=language_to_cartesia_language(Language.EN),
+            language=Language.EN,
             generation_config=None,
             pronunciation_dict_id=None,
         )
@@ -320,7 +320,7 @@ class CartesiaTTSService(WebsocketTTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.generation_config is not None:
                     default_settings.generation_config = params.generation_config
                 if params.pronunciation_dict_id is not None:
@@ -749,7 +749,7 @@ class CartesiaHttpTTSService(TTSService):
         default_settings = self.Settings(
             model="sonic-3",
             voice=None,
-            language=language_to_cartesia_language(Language.EN),
+            language=Language.EN,
             generation_config=None,
             pronunciation_dict_id=None,
         )
@@ -767,7 +767,7 @@ class CartesiaHttpTTSService(TTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.generation_config is not None:
                     default_settings.generation_config = params.generation_config
                 if params.pronunciation_dict_id is not None:
diff --git a/src/pipecat/services/elevenlabs/stt.py b/src/pipecat/services/elevenlabs/stt.py
index daca9be3d..aa7fd0659 100644
--- a/src/pipecat/services/elevenlabs/stt.py
+++ b/src/pipecat/services/elevenlabs/stt.py
@@ -272,7 +272,7 @@ class ElevenLabsSTTService(SegmentedSTTService):
         # 1. Initialize default_settings with hardcoded defaults
         default_settings = self.Settings(
             model="scribe_v2",
-            language=language_to_elevenlabs_language(Language.EN),
+            language=Language.EN,
             tag_audio_events=None,
         )
 
@@ -286,7 +286,7 @@ class ElevenLabsSTTService(SegmentedSTTService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = language_to_elevenlabs_language(params.language)
+                    default_settings.language = params.language
                 default_settings.tag_audio_events = params.tag_audio_events
 
         # 4. Apply settings delta (canonical API, always wins)
diff --git a/src/pipecat/services/elevenlabs/tts.py b/src/pipecat/services/elevenlabs/tts.py
index 8cfd1abe2..866d0405f 100644
--- a/src/pipecat/services/elevenlabs/tts.py
+++ b/src/pipecat/services/elevenlabs/tts.py
@@ -449,7 +449,7 @@ class ElevenLabsTTSService(WebsocketTTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.stability is not None:
                     default_settings.stability = params.stability
                 if params.similarity_boost is not None:
@@ -1014,7 +1014,7 @@ class ElevenLabsHttpTTSService(TTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.optimize_streaming_latency is not None:
                     default_settings.optimize_streaming_latency = params.optimize_streaming_latency
                 if params.stability is not None:
diff --git a/src/pipecat/services/fal/stt.py b/src/pipecat/services/fal/stt.py
index 7bfcfbcfd..65df7e3ab 100644
--- a/src/pipecat/services/fal/stt.py
+++ b/src/pipecat/services/fal/stt.py
@@ -216,7 +216,7 @@ class FalSTTService(SegmentedSTTService):
         # 1. Initialize default_settings with hardcoded defaults
         default_settings = self.Settings(
             model=None,
-            language=language_to_fal_language(Language.EN),
+            language=Language.EN,
         )
 
         # 2. (no deprecated direct args for this service)
@@ -226,7 +226,7 @@ class FalSTTService(SegmentedSTTService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = language_to_fal_language(params.language)
+                    default_settings.language = params.language
                 if params.task != "transcribe":
                     task = params.task
                 if params.chunk_level != "segment":
diff --git a/src/pipecat/services/google/tts.py b/src/pipecat/services/google/tts.py
index 3455a8125..93053cc94 100644
--- a/src/pipecat/services/google/tts.py
+++ b/src/pipecat/services/google/tts.py
@@ -653,7 +653,7 @@ class GoogleHttpTTSService(TTSService):
                 if params.emphasis is not None:
                     default_settings.emphasis = params.emphasis
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.gender is not None:
                     default_settings.gender = params.gender
                 if params.google_style is not None:
@@ -1090,7 +1090,7 @@ class GoogleTTSService(GoogleBaseTTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.speaking_rate is not None:
                     default_settings.speaking_rate = params.speaking_rate
 
@@ -1346,7 +1346,7 @@ class GeminiTTSService(GoogleBaseTTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.prompt is not None:
                     default_settings.prompt = params.prompt
                 if params.multi_speaker is not None:
diff --git a/src/pipecat/services/groq/stt.py b/src/pipecat/services/groq/stt.py
index 6ff4075ae..3f6c23774 100644
--- a/src/pipecat/services/groq/stt.py
+++ b/src/pipecat/services/groq/stt.py
@@ -85,7 +85,7 @@ class GroqSTTService(BaseWhisperSTTService):
         # --- 1. Hardcoded defaults ---
         default_settings = self.Settings(
             model="whisper-large-v3-turbo",
-            language=self.language_to_service_language(Language.EN),
+            language=Language.EN,
             prompt=None,
             temperature=None,
         )
@@ -96,7 +96,7 @@ class GroqSTTService(BaseWhisperSTTService):
             default_settings.model = model
         if language is not None:
             self._warn_init_param_moved_to_settings("language", "language")
-            default_settings.language = self.language_to_service_language(language)
+            default_settings.language = language
         if prompt is not None:
             self._warn_init_param_moved_to_settings("prompt", "prompt")
             default_settings.prompt = prompt
diff --git a/src/pipecat/services/kokoro/tts.py b/src/pipecat/services/kokoro/tts.py
index 34e703dab..4756d4e74 100644
--- a/src/pipecat/services/kokoro/tts.py
+++ b/src/pipecat/services/kokoro/tts.py
@@ -150,7 +150,7 @@ class KokoroTTSService(TTSService):
         default_settings = self.Settings(
             model=None,
             voice=None,
-            language=language_to_kokoro_language(Language.EN),
+            language=Language.EN,
         )
 
         # 2. Apply direct init arg overrides (deprecated)
@@ -162,7 +162,7 @@ class KokoroTTSService(TTSService):
         if params is not None:
             self._warn_init_param_moved_to_settings("params")
             if not settings:
-                default_settings.language = language_to_kokoro_language(params.language)
+                default_settings.language = params.language
 
         # 4. Apply settings delta (canonical API, always wins)
         if settings is not None:
diff --git a/src/pipecat/services/lmnt/tts.py b/src/pipecat/services/lmnt/tts.py
index 0ca91a107..5fe574c5a 100644
--- a/src/pipecat/services/lmnt/tts.py
+++ b/src/pipecat/services/lmnt/tts.py
@@ -129,7 +129,7 @@ class LmntTTSService(InterruptibleTTSService):
         default_settings = self.Settings(
             model="aurora",
             voice=None,
-            language=self.language_to_service_language(language),
+            language=language,
         )
 
         # 2. Apply direct init arg overrides (deprecated)
diff --git a/src/pipecat/services/neuphonic/tts.py b/src/pipecat/services/neuphonic/tts.py
index b345a0ff4..6fb9d3e28 100644
--- a/src/pipecat/services/neuphonic/tts.py
+++ b/src/pipecat/services/neuphonic/tts.py
@@ -153,7 +153,7 @@ class NeuphonicTTSService(InterruptibleTTSService):
         default_settings = self.Settings(
             model=None,
             voice=None,
-            language=self.language_to_service_language(Language.EN),
+            language=Language.EN,
             speed=1.0,
         )
 
@@ -167,7 +167,7 @@ class NeuphonicTTSService(InterruptibleTTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.speed is not None:
                     default_settings.speed = params.speed
 
@@ -487,7 +487,7 @@ class NeuphonicHttpTTSService(TTSService):
         default_settings = self.Settings(
             model=None,
             voice=None,
-            language=self.language_to_service_language(Language.EN),
+            language=Language.EN,
             speed=1.0,
         )
 
@@ -501,7 +501,7 @@ class NeuphonicHttpTTSService(TTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = self.language_to_service_language(params.language)
+                    default_settings.language = params.language
                 if params.speed is not None:
                     default_settings.speed = params.speed
 
diff --git a/src/pipecat/services/nvidia/stt.py b/src/pipecat/services/nvidia/stt.py
index fcca14741..50a654191 100644
--- a/src/pipecat/services/nvidia/stt.py
+++ b/src/pipecat/services/nvidia/stt.py
@@ -503,7 +503,7 @@ class NvidiaSegmentedSTTService(SegmentedSTTService):
         # 1. Initialize default_settings with hardcoded defaults
         default_settings = self.Settings(
             model=model_function_map.get("model_name"),
-            language=language_to_nvidia_riva_language(Language.EN_US) or "en-US",
+            language=Language.EN_US,
             profanity_filter=False,
             automatic_punctuation=True,
             verbatim_transcripts=False,
@@ -517,9 +517,7 @@ class NvidiaSegmentedSTTService(SegmentedSTTService):
         if params is not None:
             self._warn_init_param_moved_to_settings("params")
             if not settings:
-                default_settings.language = (
-                    language_to_nvidia_riva_language(params.language or Language.EN_US) or "en-US"
-                )
+                default_settings.language = params.language or Language.EN_US
                 default_settings.profanity_filter = params.profanity_filter
                 default_settings.automatic_punctuation = params.automatic_punctuation
                 default_settings.verbatim_transcripts = params.verbatim_transcripts
diff --git a/src/pipecat/services/openai/stt.py b/src/pipecat/services/openai/stt.py
index 39ca60b25..68ab7e067 100644
--- a/src/pipecat/services/openai/stt.py
+++ b/src/pipecat/services/openai/stt.py
@@ -119,7 +119,7 @@ class OpenAISTTService(BaseWhisperSTTService):
         _language = language or Language.EN
         default_settings = self.Settings(
             model="gpt-4o-transcribe",
-            language=self.language_to_service_language(_language),
+            language=_language,
             prompt=None,
             temperature=None,
         )
diff --git a/src/pipecat/services/rime/tts.py b/src/pipecat/services/rime/tts.py
index 24a5d152b..aec5630b6 100644
--- a/src/pipecat/services/rime/tts.py
+++ b/src/pipecat/services/rime/tts.py
@@ -251,9 +251,7 @@ class RimeTTSService(WebsocketTTSService):
         if params is not None:
             self._warn_init_param_moved_to_settings("params")
             if not settings:
-                default_settings.language = (
-                    self.language_to_service_language(params.language) if params.language else None
-                )
+                default_settings.language = params.language
                 default_settings.segment = params.segment
                 default_settings.speedAlpha = params.speed_alpha
                 # Arcana params
@@ -754,9 +752,7 @@ class RimeHttpTTSService(TTSService):
         if params is not None:
             self._warn_init_param_moved_to_settings("params")
             if not settings:
-                default_settings.language = (
-                    self.language_to_service_language(params.language) if params.language else "eng"
-                )
+                default_settings.language = params.language
                 default_settings.speedAlpha = params.speed_alpha
                 default_settings.reduceLatency = params.reduce_latency
                 default_settings.pauseBetweenBrackets = params.pause_between_brackets
@@ -984,9 +980,7 @@ class RimeNonJsonTTSService(InterruptibleTTSService):
         if params is not None:
             self._warn_init_param_moved_to_settings("params")
             if not settings:
-                default_settings.language = (
-                    self.language_to_service_language(params.language) if params.language else None
-                )
+                default_settings.language = params.language
                 default_settings.segment = params.segment
                 default_settings.repetition_penalty = params.repetition_penalty
                 default_settings.temperature = params.temperature
diff --git a/src/pipecat/services/sambanova/stt.py b/src/pipecat/services/sambanova/stt.py
index d3a77b4eb..5cf12d771 100644
--- a/src/pipecat/services/sambanova/stt.py
+++ b/src/pipecat/services/sambanova/stt.py
@@ -82,7 +82,7 @@ class SambaNovaSTTService(BaseWhisperSTTService):  # type: ignore
         # --- 1. Hardcoded defaults ---
         default_settings = self.Settings(
             model="Whisper-Large-v3",
-            language=self.language_to_service_language(Language.EN),
+            language=Language.EN,
             prompt=None,
             temperature=None,
         )
@@ -93,7 +93,7 @@ class SambaNovaSTTService(BaseWhisperSTTService):  # type: ignore
             default_settings.model = model
         if language is not None:
             self._warn_init_param_moved_to_settings("language", "language")
-            default_settings.language = self.language_to_service_language(language)
+            default_settings.language = language
         if prompt is not None:
             self._warn_init_param_moved_to_settings("prompt", "prompt")
             default_settings.prompt = prompt
diff --git a/src/pipecat/services/sarvam/tts.py b/src/pipecat/services/sarvam/tts.py
index 91ee088cf..c926270de 100644
--- a/src/pipecat/services/sarvam/tts.py
+++ b/src/pipecat/services/sarvam/tts.py
@@ -473,9 +473,7 @@ class SarvamHttpTTSService(TTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = (
-                        self.language_to_service_language(params.language) or "en-IN"
-                    )
+                    default_settings.language = params.language
                 if params.enable_preprocessing is not None:
                     default_settings.enable_preprocessing = params.enable_preprocessing
                 if params.pace is not None:
@@ -491,10 +489,6 @@ class SarvamHttpTTSService(TTSService):
         if settings is not None:
             default_settings.apply_update(settings)
 
-        # Convert Language enum to service-specific string
-        if isinstance(default_settings.language, Language):
-            default_settings.language = self.language_to_service_language(default_settings.language)
-
         # Get model configuration (validates model exists)
         resolved_model = default_settings.model
         if resolved_model not in TTS_MODEL_CONFIGS:
@@ -889,9 +883,7 @@ class SarvamTTSService(InterruptibleTTSService):
             self._warn_init_param_moved_to_settings("params")
             if not settings:
                 if params.language is not None:
-                    default_settings.language = (
-                        self.language_to_service_language(params.language) or "en-IN"
-                    )
+                    default_settings.language = params.language
                 if params.enable_preprocessing is not None:
                     default_settings.enable_preprocessing = params.enable_preprocessing
                 if params.min_buffer_size is not None:
@@ -915,10 +907,6 @@ class SarvamTTSService(InterruptibleTTSService):
         if settings is not None:
             default_settings.apply_update(settings)
 
-        # Convert Language enum to service-specific string
-        if isinstance(default_settings.language, Language):
-            default_settings.language = self.language_to_service_language(default_settings.language)
-
         # Get model configuration (validates model exists)
         resolved_model = default_settings.model
         if resolved_model not in TTS_MODEL_CONFIGS:
diff --git a/src/pipecat/services/whisper/base_stt.py b/src/pipecat/services/whisper/base_stt.py
index 93bb5de34..33d19b0aa 100644
--- a/src/pipecat/services/whisper/base_stt.py
+++ b/src/pipecat/services/whisper/base_stt.py
@@ -194,7 +194,7 @@ class BaseWhisperSTTService(SegmentedSTTService):
             default_settings.model = model
         if language is not None:
             self._warn_init_param_moved_to_settings("language", "language")
-            default_settings.language = self.language_to_service_language(language)
+            default_settings.language = language
         if prompt is not None:
             self._warn_init_param_moved_to_settings("prompt", "prompt")
             default_settings.prompt = prompt
diff --git a/src/pipecat/services/xtts/tts.py b/src/pipecat/services/xtts/tts.py
index 099fce65c..1c3feedb6 100644
--- a/src/pipecat/services/xtts/tts.py
+++ b/src/pipecat/services/xtts/tts.py
@@ -117,7 +117,7 @@ class XTTSService(TTSService):
         default_settings = self.Settings(
             model=None,
             voice=None,
-            language=self.language_to_service_language(language),
+            language=language,
         )
 
         # 2. Apply direct init arg overrides (deprecated)