Technical implementation details and architecture for voicemail detection
text
to ensure text-based responsesgpt-4o
, GoogleLLMService with gemini-2.0-flash
voicemail_response_delay
parameter should be tuned based on your target voicemail systems. The default value of 2 seconds is a good starting point, but you may need to adjust it based on your target voicemail systems.
detector()
and gate()
are correctly placed in pipelinevoicemail_response_delay
based on observed voicemail greeting patternsgate()
is placed immediately after TTS serviceCLASSIFIER_RESPONSE_INSTRUCTION
constant to ensure proper functionality.