The Voicemagic REST API lets you generate speech, manage voices, and retrieve generations programmatically. It also exposes an MCP server for AI agent integrations.
https://voicemagic.devAll API requests require an API key passed as a Bearer token in the Authorization header. Generate a key in your Settings page.
Authorization: Bearer vm_your_api_key_here
API keys are scoped to your organization. They grant access to your organization's custom voices and generations. Store them securely — you will only see the full key once when it is created.
/api/v1/voicesList all available voices — both system voices shared across all organizations, and custom voices you have created.
Response
{
"voices": [
{
"id": "clx...",
"name": "Sarah",
"description": "Warm and conversational",
"category": "CONVERSATIONAL",
"language": "en-US",
"variant": "SYSTEM"
}
]
}cURL example
curl https://voicemagic.dev/api/v1/voices \ -H "Authorization: Bearer vm_your_api_key"
/api/v1/ttsGenerate speech from text. Returns a generation ID and an audio URL. Consumes metered credits.
Request body
{
"text": "Hello from Voicemagic!",
"voiceId": "clx...",
"temperature": 0.8,
"topP": 0.95,
"topK": 1000,
"repetitionPenalty": 1.2
}Response
{
"id": "clx...",
"audioUrl": "/api/v1/audio/clx..."
}cURL example
curl -X POST https://voicemagic.dev/api/v1/tts \
-H "Authorization: Bearer vm_your_api_key" \
-H "Content-Type: application/json" \
-d '{
"text": "Hello from Voicemagic!",
"voiceId": "clx..."
}'/api/v1/audio/{id}Stream the WAV audio for a generation by its ID. Returns the raw audio bytes with Content-Type: audio/wav.
Response
<binary WAV audio stream>
cURL example
curl https://voicemagic.dev/api/v1/audio/clx... \ -H "Authorization: Bearer vm_your_api_key" \ --output speech.wav
/api/v1/generations/{id}Retrieve metadata and the audio URL for a specific generation.
Response
{
"id": "clx...",
"text": "Hello from Voicemagic!",
"voiceName": "Sarah",
"voiceId": "clx...",
"temperature": 0.8,
"topP": 0.95,
"topK": 1000,
"repetitionPenalty": 1.2,
"createdAt": "2026-04-21T12:00:00.000Z",
"audioUrl": "/api/v1/audio/clx..."
}cURL example
curl https://voicemagic.dev/api/v1/generations/clx... \ -H "Authorization: Bearer vm_your_api_key"
Voicemagic exposes a Model Context Protocol (MCP) server, allowing AI agents like Claude to generate speech directly.
Endpoint
https://voicemagic.dev/api/mcp
Available tools
list_voicesLists all available system and custom voices for your organization.
generate_speechGenerates TTS audio from text using a specified voice. Returns an audio URL.
get_generationRetrieves generation metadata and the audio URL for a given generation ID.
Add to Claude Code (claude_code_config.json)
{
"mcpServers": {
"voicemagic": {
"type": "http",
"url": "https://voicemagic.dev/api/mcp",
"headers": {
"Authorization": "Bearer vm_your_api_key"
}
}
}
}