narrador API & MCP server reference — text to speech for AI agents

narrador is a public REST API and remote MCP server for text to speech aimed at AI agents: narrador files turns documents into audio, and narrador X turns X/Twitter posts into audio.

Product site: https://narrador.io

Authentication

Send a user API key on every request using either header:

Authorization: Bearer <key>
x-api-key: <key>

Keys are scoped by prefix: nrd_files_ for narrador files, nrd_x_ for narrador X.

MCP

Remote MCP endpoint (Streamable HTTP): https://api.narrador.io/mcp

Tools: get_account, list_voices, create_files_job, get_files_job, get_files_result, generate_x_audio, get_x_generation

REST endpoints

GET /v1/me — Current user and credits
GET /v1/voices — List voices
POST /v1/files/uploads — Create upload URL
POST /v1/files/jobs — Start an audio job
GET /v1/files/jobs/{jobId} — Job status
GET /v1/files/jobs/{jobId}/result — Job result
POST /v1/x/generations — Generate post audio
GET /v1/x/generations/{generationId} — Get generation result