elevenlabs new websocket connection on each turn #824
Description
Describe the bug
The current elevenlabs tts implementation is creating a new websocket connection on each turn.
We should implement the https://elevenlabs.io/docs/api-reference/text-to-speech/v-1-text-to-speech-voice-id-multi-stream-input like the python version does.
Relevant log output
No response
Describe your environment
Minimal reproducible example
No response
Additional information
No response