Gemini CLI reaches your Speak AI account through the MCP server, so it can read your recordings, transcripts and insights while you work.
What you need
A Speak AI account, and Gemini CLI. Some setup paths also need an API key, which you create at app.speakai.co/developers/apikeys.
Every client connects to the same address:
https://api.speakai.co/v1/mcpHow do you connect Gemini CLI?
-
Add the connector:
gemini mcp add --transport http speakai https://api.speakai.co/v1/mcp -
Start Gemini CLI and type
/mcp auth speakai -
Your browser opens Speak AI’s sign-in page. Sign in, or create an account if you don’t have one, then click Allow
-
Sign-in needs a browser on the same machine. Over SSH, use an API key instead (below)
What can you ask for once it is connected?
Ask in your own words. The server has tools for media, transcripts, AI insights, folders, recorders, automations and more, so most questions about your account resolve without naming one:
- “Summarise my last customer interview.”
- “Find every mention of pricing across this month’s calls.”
- “Upload this recording and tell me the main themes.”
Related guides
- Connect your AI tool covers every supported client.
- Connect Claude.ai (web) to Speak AI
- Speak AI MCP server explains what the server exposes.
- MCP authentication and rate limits