Give your apps and AI agents access to DubSmart

Automate AI dubbing, text-to-speech, voice cloning, transcription, captions, image-to-video, and project management through a remote MCP server or the REST API.

Choose MCP for agents

MCP exposes 75 tools that agents can discover and call. Use it when a compatible client should plan and run complete DubSmart workflows on a user’s behalf.

Choose REST for software

The REST API gives your backend explicit HTTP endpoints, JSON responses, bearer authentication, and an OpenAPI description for generated clients.

Connect an MCP client

  1. 1

    Create a token

    Sign in, open API & MCP tokens, choose Create MCP token, and select the narrowest access preset that fits the workflow.

  2. 2

    Save the secret

    The dsmcp_ token is shown once. Store it in the client’s secure secret storage; do not paste it into prompts or source control.

  3. 3

    Add the remote server

    Use the Streamable HTTP endpoint and pass the token as a Bearer authorization header.

  4. 4

    Review sensitive actions

    Credit-consuming and sensitive tools ask for a short-lived confirmation before execution.

{
  "mcpServers": {
    "dubsmart": {
      "url": "https://mcp.dubsmart.ai/mcp",
      "headers": {
        "Authorization": "Bearer dsmcp_YOUR_TOKEN"
      }
    }
  }
}

This configuration works in clients that support remote Streamable HTTP servers and custom headers. OAuth-based one-click connectors are not part of the current release.

Control what an agent can do

MCP tokens expire after 90 days and can be revoked at any time. Start with Creator unless the agent only needs to inspect data or explicitly needs account-wide actions.
PresetBest forAccess
Read-onlyReporting and monitoringAccount, projects, voices, billing, referrals, and support reads
CreatorContent productionRead access plus project, media, voice, account, and support writes
Full accessTrusted account administrationCreator access plus deletes, sharing, referrals, and token management

Call the REST API

Create an API key on the same dashboard page, then send it in the Authorization header. The example below lists English voices; use the interactive reference for request and response schemas across all endpoints.
curl -X POST "https://dubsmart.ai/api/v1/voices" \
  -H "Authorization: Bearer dsapi_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"limit": 20, "filter": {"language": "en"}}'