ListenHub

ListenHub MCP Server lets you run the ListenHub MCP backend locally or connect to a remote MCP endpoint. It enables podcast generation, FlowSpeech creation, speaker management, and user account actions through multiple transport options so you can integrate ListenHub capabilities into your own tools and workflows.

How to use

You interact with the ListenHub MCP Server through client applications or custom integrations. Choose a transport method that fits your workflow: stdio for local, in-process usage, or HTTP for remote integration. You can run the server with an inline command and then connect clients using the provided MCP endpoint. Typical usage patterns include starting the server with an API key, querying available speakers, creating podcasts (text plus audio or text only), generating FlowSpeech, and checking statuses of ongoing operations.

How to install

Prerequisites you need before starting:

• Node.js version 18 or higher
• An API key from ListenHub (LISTENHUB_API_KEY) for authenticated usage
• An MCP client or transport you plan to use (stdio or http)

Step by step install and run flow:

1. Install Node.js if it is not already installed
2. Obtain your API key from ListenHub and keep it ready
3. Install and run the MCP server using the stdio approach or the HTTP transport as needed

# Start the server in HTTP mode (example)
export LISTENHUB_API_KEY="your_api_key_here"
npx @marswave/listenhub-mcp-server --transport http --port 3000

# Available endpoints when running in HTTP mode
# HTTP Streaming: http://localhost:3000/mcp
# Server-Sent Events: http://localhost:3000/sse

Additional sections

Configuration and usage details are provided below to help you plug the server into various clients. The server supports both HTTP transport and stdio transport, with environment variables shown for required configuration. For HTTP mode, point your MCP client to the HTTP endpoint URL, and for stdio mode, use the in-process command with the appropriate arguments and environment.

Available tools

get_speakers

Retrieve a list of available speakers for podcast generation, including language and demo audio.

create_podcast

Create a podcast with full generation (text + audio) using 1-2 speakers; supports progress polling and returns audio, script, and outlines.

get_podcast_status

Query the detailed status of a previously created podcast episode.

create_podcast_text_only

Create a text-only podcast as the first stage of a two-stage generation flow.

generate_podcast_audio

Generate audio for an existing podcast text content as the second stage of a two-stage process.

create_flowspeech

Convert text or URL content to FlowSpeech with selectable narrator and mode.

get_flowspeech_status

Query the status of a FlowSpeech generation.

get_user_subscription

Query current user subscription details and credit usage.