speak-to-me-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| OPENAI_API_KEY | No | Optional API key for OpenAI TTS premium voices. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| speakA | Speak text aloud using text-to-speech. Use this to give Claude a voice — summarize changes, announce completions, or narrate explanations audibly. Audio plays on the user's machine and the tool returns immediately without waiting for speech to finish. Default engine is macOS system TTS (free, offline). Set engine to "openai" for higher quality (requires OPENAI_API_KEY env var). |
| list_voicesB | List available text-to-speech voices. Returns system (macOS) voices and/or OpenAI voices depending on the engine parameter. |
| configure_speechA | Set default speech preferences for this session. Changes persist until the MCP server restarts. Call with no arguments to view current config. |
| speech_styleA | Set a speaking style or persona that shapes how Claude composes spoken text. This does NOT change the TTS voice — it changes the way summaries are written before being spoken aloud. Examples: "talk like a clown", "speak like a business woman", "be super concise and direct", "explain things like a friendly teacher". Call with an empty style to clear it. Call with no arguments to view the current style. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
speak, list_voices, configure_speech, and speech_style each target fairly distinct actions, and descriptions clarify that speech_style affects text composition rather than TTS voice. However, configure_speech and speech_style both deal with settings/preferences and could initially be confused, even if the descriptions distinguish them.
Three tools follow a verb/noun or verb_object pattern (speak, list_voices, configure_speech), but speech_style breaks the imperative pattern as a noun phrase. The deviation is minor and the names remain readable.
Four tools is well-scoped for a text-to-speech server: one to speak, one to enumerate voices, and two to configure behavior. Nothing feels extraneous or missing at the count level.
The surface covers the core lifecycle of speaking, voice discovery, and configuration. A stop/interrupt or queue-control tool is absent, but the fire-and-forget design makes that a minor, workable gap.