wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(elevenlabs tts): Fix message paramter naming and make use of contexts to send out TTSStoppedFrames()

Browse Source
This commit is contained in:
Severin Klingler
2025-05-19 15:22:13 +02:00
parent 7a8ab9a900
commit e1299d59bf
1 changed files with 17 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
src/pipecat/services/elevenlabs/tts.py
View File

@@ -276,8 +276,16 @@ class ElevenLabsTTSService(AudioContextWordTTSService):
async def flush_audio(self): async def flush_audio(self):
if self._websocket and self._context_id: if self._websocket and self._context_id:
msg = {"context_id": self._context_id, "flush": True} self._context_id_to_close = self._context_id
msg = {"context_id": self._context_id, "flush": True, "xi_api_key": self._api_key}
await self._websocket.send(json.dumps(msg)) await self._websocket.send(json.dumps(msg))
msg = {
"context_id": self._context_id,
"close_context": True,
"xi_api_key": self._api_key,
}
await self._websocket.send(json.dumps(msg))
self._context_id = None
async def push_frame(self, frame: Frame, direction: FrameDirection = FrameDirection.DOWNSTREAM): async def push_frame(self, frame: Frame, direction: FrameDirection = FrameDirection.DOWNSTREAM):
await super().push_frame(frame, direction) await super().push_frame(frame, direction)
@@ -386,7 +394,7 @@ class ElevenLabsTTSService(AudioContextWordTTSService):
msg = json.loads(message) msg = json.loads(message)
# Check if this message belongs to the current context # Check if this message belongs to the current context
# The default context may return null/None for context_id # The default context may return null/None for context_id
received_ctx_id = msg.get("context_id") received_ctx_id = msg.get("contextId")
if ( if (
self._context_id is not None self._context_id is not None
and received_ctx_id is not None and received_ctx_id is not None
@@ -406,12 +414,17 @@ class ElevenLabsTTSService(AudioContextWordTTSService):
word_times = calculate_word_times(msg["alignment"], self._cumulative_time) word_times = calculate_word_times(msg["alignment"], self._cumulative_time)
await self.add_word_timestamps(word_times) await self.add_word_timestamps(word_times)
self._cumulative_time = word_times[-1][1] self._cumulative_time = word_times[-1][1]
if msg.get("is_final"): if msg.get("isFinal"):
logger.trace(f"Received final message for context {received_ctx_id}") logger.trace(f"Received final message for context {received_ctx_id}")
# Context has finished # Context has finished
if self._context_id == received_ctx_id: if (
self._context_id == received_ctx_id
or self._context_id_to_close == received_ctx_id
):
self._context_id = None self._context_id = None
self._context_id_to_close = None
self._started = False self._started = False
await self.push_frame(TTSStoppedFrame())
async def _keepalive_task_handler(self): async def _keepalive_task_handler(self):
while True: while True: