processors(realtime-ai): fix voice initialization
This commit is contained in:
@@ -166,7 +166,7 @@ class RealtimeAIProcessor(FrameProcessor):
|
||||
|
||||
voice = DEFAULT_VOICE
|
||||
if setup and setup.config and setup.config.tts and setup.config.tts.voice:
|
||||
messages = setup.config.tts.voice
|
||||
voice = setup.config.tts.voice
|
||||
|
||||
self._tma_in = LLMUserResponseAggregator(messages)
|
||||
self._tma_out = LLMAssistantResponseAggregator(messages)
|
||||
|
||||
Reference in New Issue
Block a user