kokoro-tts-mcpKokoro Text to Speech (TTS) MCP Server

— (0)

Reviewed by Daniel Nikulshyn·Updated July 2026

Overview

The Kokoro Text to Speech (TTS) MCP Server is an open-source Text-to-Speech system that generates .mp3 files with an option to upload them to S3. It uses the Kokoro-TTS model provided by Hugging Face. The system allows users to customize voices, speed, and language. The server is configured through environment variables in the .env file. Supported environment variables include AWS access keys, S3 bucket information, and debug mode. The server can be run locally using the UV library. The TTS client, found in the mcp_client.py script, allows users to send TTS requests to the server. It supports various options, such as providing text, reading from a file, customizing voice and speed, and disabling S3 upload. Local MP3 files are stored in a specified folder, and users can configure automatic cleanup by setting the number of days to retain files and whether to delete local files after successful S3 upload.

Key features

Text-to-speech functionality
Customizable voice, speed, and language
S3 upload integration
Automatic MP3 file management
Local server and client script for management

Pricing

Model: Free
Category: MCP Servers
Rating: No reviews yet

Use cases

Content Creation

Generate audio content from text for podcasts, audiobooks, or social media using customizable voices and speeds.

Automation

Automatically generate and upload MP3 files to S3 for use in applications or websites.

Pros & Cons

Pros

Customizable TTS voices and speeds
Optional S3 upload for generated MP3 files
Automatic cleanup of local MP3 files

Cons

Requires configuration of environment variables and MCP settings
Dependent on additional tools like ffmpeg for .wav to .mp3 conversion

Reviews

No reviews yet. Be the first!

Q&A

What are the limitations of running this TTS server locally?

You need to configure several environment variables, install ffmpeg for conversion, and manage local MP3 storage. Automatic cleanup is based on MP3_RETENTION_DAYS, and deleting after upload is optional. The system requires Python, uv, and the Hugging Face model files to be downloaded locally.

Asked by Grzegorz Lewandowski · Feb 9, 2026

How do I customize voice, speed, and language in the TTS output?

Set the environment variables TTS_VOICE, TTS_SPEED, and TTS_LANGUAGE in the .env file or pass them as options when calling mcp_client.py. The server uses the Kokoro‑TTS model from Hugging Face, which supports multiple voices and languages.

Asked by Ines Zeković · Jan 28, 2026

What are the main integrations that Kokoro TTS supports?

The server integrates with Amazon S3 for optional MP3 uploads, requires ffmpeg for converting .wav to .mp3, and can be accessed via a local client script (mcp_client.py) or any HTTP client using the MCP configuration.

Asked by Katarzyna Zielinska · Dec 21, 2025

Is Kokoro TTS MCP Server free to use or is there a licensing cost?

Kokoro TTS MCP Server is open-source and released under a license that allows free use. No subscription or per‑use fees are required, but you must host and maintain the server yourself.

Asked by Camille Laurent · Dec 15, 2025