mcpField
V
voize

voize

github.com/zenon-redCreative & media259 installsUpdated 22w ago
Repository

About voize

Voize is a local MCP server for generating speech and returning a public URL to the uploaded audio. It currently uses Xiaomi MiMo for TTS and Cloudflare R2 for storage.

Key features

  1. 01Speech synthesis with Xiaomi MiMo
  2. 02Upload audio to Cloudflare R2
  3. 03Return a public audio URL
  4. 04Support built-in and cloned voices

Use cases

  1. 01Generate speech audio for an application
  2. 02Create audio using a voice sample
  3. 03Make generated audio available through a public URL

MCP support

Voize runs as a local stdio MCP server and exposes the `generate_tts_url` tool. The tool synthesizes speech, uploads the audio, and returns a public URL.

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/zenon-red/skills --skill voize

After installing, type/voizein Claude Code to confirm it loaded.

Bundled filesMETA.md

Details

From the project README

Voize

Overview

Voize is a local MCP server that provides one tool, generate_tts_url.

It synthesizes speech using the current backend, uploads bytes to the current storage backend, and returns a public URL.

Current concrete support is Xiaomi MiMo (TTS) + Cloudflare R2 (storage), while keeping extension seams for future backends.

Tech Stack

  • Language: TypeScript
  • Runtime: Bun 1.0+
  • Key dependencies: @modelcontextprotocol/sdk, @aws-sdk/client-s3, zod

Architecture

The server is stdio MCP with one tool handler.

src/
├── index.ts
├── lib/
│   ├── provider.ts
│   └── storage.ts
├── tools/
│   └── generateTtsUrl.ts
├── types.ts
└── utils/
    ├── env.ts
    ├── log.ts
    ├── retry.ts
    └── validate.ts

Development

Setup

Fill required env vars and enforce public-domain guardrails.

bun install
cp .env.example .env
Build

bun run build

Test

No automated tests yet; validate by MCP smoke call with built-in and voice-clone inputs.

Lint

bun run lint

Typecheck

bun run typecheck

Key Files

FilePurpose
src/index.tsMCP bootstrap, env validation, stdio transport
src/tools/generateTtsUrl.tsInput validation, TTS call, upload, MCP response
src/lib/provider.tsCurrent TTS backend client
src/lib/storage.tsCurrent S3-compatible storage helper
docs/architecture.mdEnd-to-end flow and constraints

Agent Guidelines

  • Never write to stdout in server code (stdio JSON-RPC safety).
  • Return structured JSON in MCP text content for both success and failure.
  • Keep public-domain enforcement at startup.
  • Do not introduce direct Nexus/STDB calls; keep boundary at URL generation.
  • Hosted reference voice sample for cloning workflows: https://audio.zenon.red/voice/samples/zoe-sample.mp3.
  • For clone calls, pass voiceSampleUrl directly, or download and pass voiceSamplePath.

Comments

Markdown supported