From 1bbdf02618cbebe95799b0b72908a459d0b8d735 Mon Sep 17 00:00:00 2001
From: lauridsdev <65713140+lauridsdev@users.noreply.github.com>
Date: Sat, 14 Sep 2024 21:51:55 +0200
Subject: [PATCH] add voice control for speed and emotions

---
 src/pipecat/services/cartesia.py | 28 ++++++++++++++++++++++++++--
 1 file changed, 26 insertions(+), 2 deletions(-)

diff --git a/src/pipecat/services/cartesia.py b/src/pipecat/services/cartesia.py
index ac02ea469..aa27cb00b 100644
--- a/src/pipecat/services/cartesia.py
+++ b/src/pipecat/services/cartesia.py
@@ -102,6 +102,10 @@ def __init__(
         self._context_id = None
         self._receive_task = None
 
+        # Initialize default values for speed and emotion controls
+        self._speed = "normal"  # default speed
+        self._emotion = []  # default to no emotion
+
     def can_generate_metrics(self) -> bool:
         return True
 
@@ -200,8 +204,19 @@ async def _receive_task_handler(self):
         except Exception as e:
             logger.exception(f"{self} exception: {e}")
 
+        # Method to set speed and emotion externally
+    def set_voice_controls(self, speed: str = "normal", emotion: list[str] = None):
+        """
+        Set the voice controls for TTS (speed and emotion).
+        """
+        self._speed = speed
+        self._emotion = emotion or []
+
     async def run_tts(self, text: str) -> AsyncGenerator[Frame, None]:
-        logger.debug(f"Generating TTS: [{text}]")
+        """
+        Generate TTS using the configured speed and emotion settings.
+        """
+        logger.debug(f"Generating TTS: [{text}], Speed: [{self._speed}], Emotion: [{self._emotion}]")
 
         try:
             if not self._websocket:
@@ -212,6 +227,12 @@ async def run_tts(self, text: str) -> AsyncGenerator[Frame, None]:
                 await self.start_ttfb_metrics()
                 self._context_id = str(uuid.uuid4())
 
+            # Define the voice control parameters using stored instance variables
+            voice_controls = {
+                "speed": self._speed,
+                "emotion": self._emotion
+            }
+
             msg = {
                 "transcript": text + " ",
                 "continue": True,
@@ -224,7 +245,9 @@ async def run_tts(self, text: str) -> AsyncGenerator[Frame, None]:
                 "output_format": self._output_format,
                 "language": self._language,
                 "add_timestamps": True,
+                "__experimental_controls": voice_controls  # Add the voice controls
             }
+
             try:
                 await self._websocket.send(json.dumps(msg))
                 await self.start_tts_usage_metrics(text)
@@ -234,6 +257,7 @@ async def run_tts(self, text: str) -> AsyncGenerator[Frame, None]:
                 await self._disconnect()
                 await self._connect()
                 return
+
             yield None
         except Exception as e:
-            logger.exception(f"{self} exception: {e}")
+            logger.exception(f"{self} exception: {e}")
\ No newline at end of file