Typecast API

Typecast API is an MCP server that connects MCP clients like Claude Desktop and Cursor to Neosapience's Typecast voice generation platform. Media creators, developers, and AI agent operators use this server to automate text-to-speech synthesis, manage voice libraries, and handle custom voice cloning within their development workflows. The server interfaces with Typecast models, including ssfm-v30 and ssfm-v21, supporting standard text-to-speech, streaming output, and generation with character timestamps. It provides emotional modulation through presets such as happy, sad, angry, and whisper, alongside a Smart Mode that automatically infers tone using surrounding context. Beyond standard narration across 37 supported languages, the server enables users to discover voices based on use-case tags, clone custom voices from WAV or MP3 audio files, initiate professional long-form clones, normalize output audio using target LUFS settings, and control local audio playback. It can be run locally or connected via a hosted HTTP endpoint.

Category: Design, Media & Creative

Tags: audio, media, text to speech, voice generation

Visit Typecast API

How to install and configure Typecast API

Follow these steps to configure the Typecast API MCP server using uvx: 1. Obtain an API key from your Typecast account. 2. Open your client configuration file, such as claude_desktop_config.json for Claude Desktop or Cursor MCP settings. 3. Add typecast-api-mcp-server under the mcpServers object: json { "mcpServers": { "typecast-api-mcp-server": { "command": "uvx", "args": [ "--from", "git+https://github.com/neosapience/typecast-api-mcp-server.git", "typecast-api-mcp-server" ], "env": { "TYPECAST_API_KEY": "YOUR_API_KEY", "TYPECAST_OUTPUT_DIR": "PATH/TO/YOUR/OUTPUT/DIR" } } } } 4. If running on Linux, include "XDG_RUNTIME_DIR": "/run/user/1000" inside the env block. 5. Save the configuration and restart your MCP client.

What you can do with Typecast API

• Generating emotional voiceovers across 37 languages using presets like whisper, happy, or sad with the ssfm-v30 model. - Inferring narration mood automatically across script segments using context-aware Smart Mode parameters. - Creating custom cloned voices directly from local MP3 or WAV audio clips under 25 MB. - Retrieving recommended voice profiles by querying specific styles, moods, or targeted use cases. - Exporting synchronized speech tracks with detailed timestamps for animated video and subtitle alignment. - Normalizing synthesized speech loudness to specified target LUFS standards for broadcast production.

Key facts

• https://github.com/neosapience/typecast-api-mcp-server-sample
• Design, Media & Creative
• audio, media, text to speech, voice generation

Part of MCP Servers

What is Typecast API?

Typecast API is an MCP server that provides tools for interacting with Typecast speech generation services. It enables language model clients to synthesize audio, clone voices, inspect custom models, and adjust emotional delivery using standardized Model Context Protocol tools.

Which MCP clients work with Typecast API?

The server functions with any client that implements the Model Context Protocol, including Claude Desktop, Cursor, and custom agent applications that support local stdio or hosted HTTP streamable endpoints.

What audio models does Typecast API support?

The server supports ssfm-v30 and ssfm-v21. The ssfm-v30 model provides seven emotion presets, Smart Mode context-based emotion detection, and support for 37 languages, while ssfm-v21 is a stable production model with preset-only controls.

Is Typecast API open source?

Yes, the repository sample code for the Typecast API MCP server is open source and distributed under the MIT License on GitHub.

Can I clone voices using Typecast API?

Yes, the server provides tools for instant voice cloning from local WAV or MP3 files up to 25 megabytes, as well as tools to initiate and check the status of professional voice cloning tasks.