Add speed and emotion options for Cartesia.

pipecat-ai · Aug 31, 2024 · 3c716f6 · 3c716f6
1 parent 7c342f7
commit 3c716f6
Showing 1 changed file with 17 additions and 4 deletions.
diff --git a/src/pipecat/services/cartesia.py b/src/pipecat/services/cartesia.py
@@ -73,6 +73,8 @@ def __init__(
             encoding: str = "pcm_s16le",
             sample_rate: int = 16000,
             language: str = "en",
+            speed: str = None,
+            emotion: list = None,
             **kwargs):
         super().__init__(**kwargs)
 
@@ -99,6 +101,8 @@ def __init__(
             "sample_rate": sample_rate,
         }
         self._language = language
+        self._speed = speed
+        self._emotion = emotion
 
         self._websocket = None
         self._context_id = None
@@ -249,15 +253,24 @@ async def run_tts(self, text: str) -> AsyncGenerator[Frame, None]:
                 await self.start_ttfb_metrics()
                 self._context_id = str(uuid.uuid4())
 
+            voice_config = {
+                "mode": "id",
+                "id": self._voice_id
+            }
+
+            if self._speed or self._emotion:
+                voice_config["__experimental_controls"] = {}
+                if self._speed:
+                    voice_config["__experimental_controls"]["speed"] = self._speed
+                if self._emotion:
+                    voice_config["__experimental_controls"]["emotion"] = self._emotion
+
             msg = {
                 "transcript": text + " ",
                 "continue": True,
                 "context_id": self._context_id,
                 "model_id": self._model_id,
-                "voice": {
-                    "mode": "id",
-                    "id": self._voice_id
-                },
+                "voice": voice_config,
                 "output_format": self._output_format,
                 "language": self._language,
                 "add_timestamps": True,