Munsit is the Arabic voice API. One key gets you natural Arabic speech, the most accurate Arabic speech recognition available, and everything in between. Here's the first call.
1
Your first call
Copy this, paste your key, run it.
curl
curl-X POST "https://api.munsit.com/api/v1/text-to-speech/faseeh-v1-preview" \-H"x-api-key: $MUNSIT_API_KEY" \-H"Content-Type: application/json" \-d'{"voice_id":"...","text":"مرحبا بك في منصت"}' \--output speech.wav
curl-X POST "https://api.munsit.com/api/v1/audio/transcribe" \-H"x-api-key: $MUNSIT_API_KEY" \-F"file=@meeting.mp3"
wscat-c"wss://api.munsit.com/api/v1/websocket/speech-to-text" \-H"x-api-key: $MUNSIT_API_KEY"# send audio chunks, read transcripts back
curl-X POST "https://api.munsit.com/api/v1/voices/clone" \-H"x-api-key: $MUNSIT_API_KEY" \-F"file=@sample.mp3" \-F"name=Layla"
Response
// audio bytes — save and play
Response
{
"statusCode": 200,
"data": {
"transcriptionId": "805059bf-7c3f-4a1e-9d2b-1f0c6ae83b47",
"transcription": "اجتماع الفريق يبدأ الساعة العاشرة"
},
"message": "Success"
}
Response
// transcripts arrive as the speaker talks
{ "type": "transcript", "transcription": "اجتماع الفريق يبدأ الساعة العاشرة" }
If you can't send audio to a cloud, you don't have to.
On-premises deployment. Run Munsit entirely inside your own data centre or private environment.Self hosting →Data privacy →
5
Build with an AI assistant
These docs are also an MCP server, so your assistant can read them itself instead of you pasting pages into a chat window.
terminal
claude mcp add --transport http munsit-docs https://docs.munsit.com/api/mcp
Five tools: search the docs, read a page, list every page, list every endpoint, and pull one endpoint's full schema. No key, no auth, nothing to install.MCP server →llms-full.txt →
Working with an AI assistant? Every page is available as Markdown: add .md to the URL, or send an Accept: text/markdown header. For the whole documentation in one request, point it at llms-full.txt; the page index is llms.txt. Or use Copy Page, top right.
