Help me build chatbot

Hey! I’m working on a project where I need to process user text input from frontend and generate agent audio output. . I’ve come across examples for text-to-text and audio-to-audio interactions in the library, but I haven’t found a clear approach for combining them into a text-to-audio conversation. Could you suggest any tool to achieve this?

Pipecat dont know how to implement text input

Flowise i dont know how to implement speech output

Voiceflow i dont know how to implement local model

Livekit ?