Skip to main content
Glama
Sign In
enesjakozh
Voicevox MCP Server

Voicevox MCP Server

by Dosugamea
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

Voicevox MCP Server

VOICEVOX互換の音声合成サーバー(AivisSpeech / VOICEVOX / COEIROINK) を MCP (Model Context Protocol) 経由で利用するためのサーバーです。 Cursor等でのClaude 3.7を使ったエージェントモードでの音声合成に利用できます。

必要条件

Windows環境

Docker環境 (WSL2)

  • Docker と Docker Compose

  • WSL2

  • VOICEVOX ENGINE等 (ローカルまたはDockerで実行)

  • sudo apt install libsdl2-dev pulseaudio-utils pulseaudio されたLinux環境

  • /mnt/wslg へのアクセス権限

Related MCP server: TTS-MCP

インストールと設定

  1. リポジトリをクローン

git clone https://github.com/Dosugamea/voicevox-mcp-server.git
cd voicevox-mcp-server

  1. 依存関係のインストール

npm install

  1. 環境変数の設定 .env_example をコピーして .env ファイルを作成し、必要に応じて設定を変更します:

VOICEVOX_API_URL=http://localhost:50021
VOICEVOX_SPEAKER_ID=1

実行方法

Windows環境での実行

エディタと別途で下記手順でサーバーを立ち上げてください。

npm run build
npm start

Docker環境での実行

エディタと別途での操作は不要です。 stdioモードで立ち上がるため直接実行することはできません。

設定方法

Windows環境での実行の場合

mcp.jsonに下記を追記してください。 接続が不安定なため切断されたら再接続してください。

        "voicevox": {
            "url": "http://localhost:10100/sse"
        }

Docker環境での実行の場合

mcp.jsonに下記を追記してください。 (作者環境での動作は確認できていません)

{
    "tools": {
        "voicevox": {
            "command": "cmd",
            "args": [
                "/c",
                "docker",
                "run",
                "-i",
                "--rm",
                "-v",
                "/mnt/wslg:/mnt/wslg",
                "-e",
                "PULSE_SERVER",
                "-e",
                "SDL_AUDIODRIVER",
                "-e",
                "VOICEVOX_API_URL",
                "-e",
                "VOICEVOX_SPEAKER_ID",
                "your-local-docker-image-name"
            ],
            "env": {
                "PULSE_SERVER": "unix:/mnt/wslg/PulseServer",
                "SDL_AUDIODRIVER": "pulseaudio",
                "VOICEVOX_API_URL": "http://host.docker.internal:50031",
                "VOICEVOX_SPEAKER_ID": "919692871"
            }
        }
    }
}

話者IDについて

話者IDは使用するVOICEVOXのモデルによって異なります。デフォルトでは「1」（四国めたん）を使用しています。 他の話者IDを使用する場合は、環境変数 VOICEVOX_SPEAKER_ID を変更してください。

話者IDの一覧は、VOICEVOX ENGINE APIの /speakers エンドポイントで確認できます。 例: curl http://localhost:50021/speakers

トラブルシューティング

  • VOICEVOXとの接続エラー: VOICEVOX ENGINEが起動していること、APIのURLが正しく設定されていることを確認してください。

  • 音声が再生されない: VLCが正しくインストールされていることと、パスが通っていることを確認してください。

  • Docker環境での音声出力問題: pulseaudioの設定が正しいか確認してください。

開発者向け情報

  • ソースコードに貢献する場合は、Issueを作成するか、Pull Requestを送信してください。

  • バグ報告や機能リクエストは、GitHubのIssue機能をご利用ください。

ライセンス

MIT License

Available Tools

1 tool
voicevoxC

VOICEVOXを使用して音声を合成し、ホストコンピュータで再生します

NameRequiredDescriptionDefault
textNo合成する文章

TDQS

C2.9/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It mentions synthesis and playback but lacks details on behavioral traits like error handling, performance characteristics (e.g., latency), or system requirements (e.g., VOICEVOX installation). This leaves significant gaps for an AI agent.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single, efficient sentence that directly states the tool's function. It is appropriately sized and front-loaded, with no unnecessary words, though it could be slightly more structured (e.g., separating synthesis and playback aspects).

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the lack of annotations and output schema, the description is incomplete. It covers the basic purpose but omits critical context such as what the tool returns (e.g., success/failure, audio data), error conditions, or usage limitations, which are essential for a synthesis/playback tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The description does not mention the 'text' parameter at all. However, schema description coverage is 100%, with the parameter documented as '合成する文章' (text to synthesize). Since the schema fully covers the parameter, the baseline score of 3 is appropriate, as the description adds no additional semantic value.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'VOICEVOXを使用して音声を合成し、ホストコンピュータで再生します' (Uses VOICEVOX to synthesize speech and play it on the host computer). It specifies the action (synthesize and play speech) and resource (VOICEVOX), but since there are no sibling tools, it doesn't need to distinguish from alternatives.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives or any contextual prerequisites. It simply states what the tool does without indicating appropriate scenarios or constraints.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 1 tool updatev1.0.0
    • First observedvoicevox
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

B3.1/5.0

Scored across 1 tool

Disambiguation5/5

With only one tool, there is no possibility of ambiguity or overlap between tools. The single tool 'voicevox' has a clear, distinct purpose for text-to-speech synthesis and playback.

Naming Consistency5/5

A single tool inherently exhibits perfect naming consistency, as there are no other tools to compare against. The tool name 'voicevox' is straightforward and matches the server's domain.

Tool Count2/5

A single tool is too few for a server's purpose, even for a focused domain like text-to-speech. This limits functionality, as agents cannot perform related operations like listing voices, adjusting parameters, or managing playback without additional tools.

Completeness2/5

The tool surface is severely incomplete for a text-to-speech domain. While the core synthesis and playback function is covered, there are obvious gaps such as fetching available voices, configuring speech parameters, pausing or stopping playback, or handling errors, which will likely cause agent failures.

Resources

Related MCP Connectors

Related MCP Servers

  • TTS-MCP

    nakamurau1
    A
    license
    Not graded
    quality
    C
    maintenance
    A Model Context Protocol server that integrates high-quality text-to-speech capabilities with Claude Desktop and other MCP-compatible clients, supporting multiple voice options and audio formats.
    11 npm
    1
    MIT

  • Bouyomi-chan MCP Server

    uraoz
    A
    license
    B
    quality
    D
    maintenance
    A Node.js server that enables AI assistants to interact with Bouyomi-chan's text-to-speech functionality through Model Context Protocol (MCP), allowing for voice reading of text with adjustable parameters.
    1
    2
    MIT

Appeared in Searches