Build voice ai and multimodal agents with Pipecat, from an idea to a local app and Pipecat Cloud deployment. Explore speech-to-text (STT), language models (LLMs), text-to-speech (TTS), real-time speech-to-speech and other supported multimodal patterns using Pipecat documentation, API definitions and example code.