{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_mt8bchafzxne","handle":"mcp-melotts","url":"https://wellknown.network/agents/mcp-melotts","links":{"self":"https://wellknown.network/agents/mcp-melotts/record.json","html":"https://wellknown.network/agents/mcp-melotts","markdown":"https://wellknown.network/agents/mcp-melotts/record.md","api":"https://wellknown.network/api/v1/agents/mcp-melotts","status":"https://wellknown.network/api/v1/agents/mcp-melotts/status","claim":"https://wellknown.network/agents/mcp-melotts/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/mcp-melotts/claim.json","badge":"https://wellknown.network/agents/mcp-melotts/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/mcp-melotts/history","tools":"https://wellknown.network/api/v1/agents/mcp-melotts/tools"},"ard":{"identifier":"urn:air::server:mcp-melotts","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"mcp-melotts","summary":"MCP server for text-to-speech using MeloTTS - generates audio from text","description":"# MCP MeloTTS 语音生成器 (MCP MeloTTS Audio Generator)\n\n本项目是一个 [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) 服务器，为 AI 助手提供将文本生成语音音频的能力，基于 MeloTTS。\n\n支持自动将长文本按标点与长度分割为不超过 100 字的小段，逐段生成 WAV 文件，并使用 ffmpeg 无损拼接为完整音频。\n\n## ✨ 功能特性\n\n- 文本转语音：使用 MeloTTS 将文本生成高质量语音\n- 自动分段：长文本自动分段，避免一次过长导致失败\n- 无损拼接：使用 ffmpeg 将分段 WAV 拼接为完整文件\n- 可配置语速、语言、设备与说话人\n- 跨平台：Windows、Linux、macOS\n\n## 🚀 安装\n\n### 使用 uvx (推荐)\n\n```bash\ndocker run -d -p 127.0.0.1:9900:8888 kisaragi29/melotts:latest\nuvx mcp-melotts@latest\n```\n\n### 使用 pip\n\n```bash\npip install mcp-melotts@latest\n```\n\n### 从源码安装\n\n```bash\ngit clone https://github.com/aardpro/mcp-melotts.git\ncd mcp-melotts\npip install -e .\n```\n\n## ⚙️ 配置\n\n### 配置\n\n#### 前置条件\n开发电脑已安装并可用：\n- ffmpeg（用于拼接音频）\n- MeloTTS（可通过本地 Python 包 melo.api 或 Docker Gradio 服务）\n\n#### 启动前检查\n启用 MCP 时会自动检查以下条件：\n- 本机 9900 端口是否启动了 MeloTTS HTTP 服务（用于 HTTP 模式）\n- 系统是否可调用 ffmpeg（用于无损拼接）\n任一不满足将抛出错误并终止启动\n\n#### 配置文件\n内置配置文件位于 src/main/config.json，可设置：\n- defaultLanguage：默认语言代码（ZH/EN/JP/ES）\n- defaultSpeed：默认语速\n- defaultDevice：默认设备（cpu 或 cuda:0）\n- defaultSpeaker：默认说话人标签\n- chunkSizeLimit：分段长度上限（推荐 100）\n\n### MCP 客户端配置\n\n将以下内容添加到你的 MCP 客户端配置中（例如 Claude Desktop, Cursor）：\n\n#### 选项 1: 使用 uvx \n\n```json\n{\n  \"mcpServers\": {\n    \"McpMeloTTS\": {\n      \"command\": \"uvx\",\n      \"args\": [\"mcp-melotts\"]\n    }\n  }\n}\n```\n\n#### 选项 2: 使用 pip\n\n```json\n{\n  \"mcpServers\": {\n    \"McpMeloTTS\": {\n      \"command\": \"mcp-melotts\"\n    }\n  }\n}\n```\n\n#### 选项 3: Windows 系统 \n\n```json\n{\n  \"mcpServers\": {\n    \"McpMeloTTS\": {\n      \"command\": \"cmd\",\n      \"args\": [\n        \"/c\",\n        \"chcp 65001 >nul && uvx mcp-melotts\"\n      ]\n    }\n  }\n}\n```\n\n#### 选项 4: Linux/macOS：\n\n```json\n{\n  \"mcpServers\": {\n    \"McpMeloTTS\": {\n      \"command\": \"python\",\n      \"args\": [\"-m\", \"main\"]\n    }\n  }\n}\n```\n\n## 🛠️ 可用工具\n\n### `mcp_melotts_generate_audio`\n\n将文本生成语音文件。适用于“根据文本生成音频文件”等请求。支持自动分段与 ffmpeg 无损拼接，支持两种调用模式：本地 melo.api 与 Docker Gradio HTTP 接口。\n\n**参数:**\n- `text` (string, 必填): 要转换为语音的文本\n- `language` (string, 可选): 语言代码（默认 ZH）\n- `speaker` (string, 可选):…","publisher":null,"homepage":"https://github.com/aardpro/mcp-melotts","repository":"https://github.com/aardpro/mcp-melotts","version":"1.0.0","license":null,"protocols":["mcp"],"tags":["audio","mcp","melo","melotts","text-to-speech","tts"],"pricing":null,"endpoints":[{"url":"pypi:mcp-melotts","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","repoUrl":"pypi","summary":"pypi","version":"pypi","description":"pypi","homepageUrl":"pypi"}},"derived":{"capabilities":[{"slug":"media.speech-synthesis","name":"Speech Synthesis","confidence":1,"provenance":"declared"},{"slug":"dev.version-control","name":"Version Control","confidence":0.745,"provenance":"derived"}],"categories":["dev","media"],"language":"es"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"mcp-melotts","registry":"pypi","observedAt":"2026-10-22T10:31:48.704Z","publishedAt":"2026-01-04T08:45:33.589451Z","latestVersion":"1.0.0"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"mcp-melotts","url":"https://pypi.org/project/mcp-melotts/","firstSeenAt":"2026-09-10T02:23:39.709Z","fetchedAt":"2026-10-22T10:29:41.333Z","normalizedAt":"2026-10-22T10:29:41.333Z"}]},"firstSeenAt":"2026-09-10T02:23:39.709Z","updatedAt":"2026-10-22T10:31:48.704Z"}