services(cartesia): allow output_format and model_id

src/pipecat/services/cartesia.py

@@ -21,11 +21,15 @@ class CartesiaTTSService(TTSService):
             *,
             api_key: str,
             voice_name: str,
+            model_id: str = "upbeat-moon",
+            output_format: str = "pcm_16000",
             **kwargs):
         super().__init__(**kwargs)
 
         self._api_key = api_key
         self._voice_name = voice_name
+        self._model_id = model_id
+        self._output_format = output_format
 
         try:
             self._client = AsyncCartesiaTTS(api_key=self._api_key)
@@ -40,14 +44,14 @@ class CartesiaTTSService(TTSService):
 
         try:
             chunk_generator = await self._client.generate(
-                transcript=text, voice=self._voice, stream=True,
-                model_id="upbeat-moon", data_rtype='array', output_format='pcm_16000',
-                # a chunk_time of 0.1 seems to be the default. there are small audio pops/gaps which
-                # we need to debug
-                chunk_time=0.1
+                stream=True,
+                transcript=text,
+                voice=self._voice,
+                model_id=self._model_id,
+                output_format=self._output_format,
             )
 
             async for chunk in chunk_generator:
-                yield AudioRawFrame(chunk['audio'], 16000, 1)
+                yield AudioRawFrame(chunk["audio"], chunk["sampling_rate"], 1)
         except Exception as e:
             logger.error(f"Cartesia exception: {e}")