j3k0/speech.sh

Overview

A minimal shell script that uses curl to send text to a text-to-speech service and play the audio aloud. It is designed for situations where a user cannot speak, such as temporary voice loss.

Best for

Best for
Developers needing a quick, scriptable text-to-speech fallback for accessibility or prototyping.

Use cases

• Quickly vocalize typed messages when voice is unavailable
• Integrate text-to-speech into shell scripts or pipelines
• Test or debug TTS endpoints with a simple curl command

How to use

Install

uv pip install --python .venv/bin/python "mcp[cli]" httpx

Tools exposed

• OPENAI_API_KEY
• SPEECH_VOICE
• SPEECH_SPEED
• SPEECH_MODEL
• SPEECH_API_URL

Tested with

Claude Desktop, Claude Code, ChatGPT

Notes

A minimal shell script that uses curl to send text to a text-to-speech service and play the audio aloud. It is designed for situations where a user cannot speak, such as temporary voice loss.

7 stars on GitHub. Last updated 2026-03-19. Licensed GPL-3.0.

Use cases

• Quickly vocalize typed messages when voice is unavailable
• Integrate text-to-speech into shell scripts or pipelines
• Test or debug TTS endpoints with a simple curl command

Pros

• Extremely lightweight and dependency-free beyond curl and a player
• Easy to understand and modify for custom TTS endpoints
• Works in any Unix-like environment with shell access

Cons

• Requires a separate audio player (e.g., aplay, afplay) to be installed
• No built-in error handling or retry logic for network failures
• Limited to basic text input; no support for SSML or voice selection

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.