# Speaking Meeting Bot
URL: /work/speaking_meeting_bot
Source: https://raw.githubusercontent.com/techwithanirudh/minimalistic-portfolio/refs/heads/main/content/work/speaking_meeting_bot.mdx

FastAPI + Pipecat meeting bots with real-time WebSocket audio and persona-driven behaviors.



Speaking Meeting Bot is a Python 3.11 FastAPI service that runs real-time meeting agents through Pipecat. It exposes REST and WebSocket endpoints for bot creation, and uses MeetingBaaS APIs to join Google Meet or Teams calls with distinct persona profiles.

Personas are stored in Markdown files so behavior, tone, and knowledge can be tuned without touching code. The audio stack combines Pipecat's pipeline with STT/TTS providers and OpenAI reasoning, while deployment tooling supports local ngrok tunnels and production base URLs.

<Tabs items="['Highlights', 'Stack']">
  <Tab value="Highlights">
    * REST + WebSocket API for spawning meeting bots.
    * Markdown-based personas with optional avatar generation.
    * Real-time audio pipelines with configurable providers.
  </Tab>

  <Tab value="Stack">
    * FastAPI, Pipecat, Python 3.11.
    * MeetingBaaS APIs + WebSocket transport.
    * Cartesia TTS, Gladia/Deepgram STT, OpenAI.
  </Tab>
</Tabs>


Last updated on April 2, 2026