Aisha API documentation
REST APIs for Uzbek speech - text-to-speech, speech-to-text and voice agents. Authenticate with an API key and make your first call in minutes.
Quickstart with the SDK #
aisha-ai is a zero-dependency SDK and CLI for Node.js (npm) and Python (pip). Install it, or run one-off commands with npx.
Node.js · npm
npm install aisha-aiPython · pip
pip install aisha-aiOne-liner
export AISHA_API_KEY=your_api_key
npx aisha-ai tts "Salom dunyo" --model Gulnoza --out salom.wavWorking with an AI agent? Point it at llms-api.txt - the whole API in one plain-text file - or ask it to install the aisha-ai package.
You can also find us on PyPI and npm.
SDKs by language
We improve our voice models every day.
TTS API #
Text-to-Speech endpoints. For server-to-server integrations, send X-Api-Key with every request.
- Send
transcripttoPOST /api/v1/tts/post/to generate audio. - Pass
webhook_notification_urlto run the request asynchronously and receive202 Accepted. - Check the result through
GET /api/v1/tts/status/{id}/or history.
Endpoints
- POST
/api/v1/tts/post/Generate audio - GET
/api/v1/tts/status/{id}/Check status - GET
/api/v1/tts/get/?page=1&limit=10History list
STT API #
Speech-to-Text endpoints. For server-to-server integrations, send X-Api-Key with each request.
-
POST /api/v1/stt/post/for short audio (sync). -
POST /api/v2/stt/post/for long audio (async, returnstask_id).
Endpoints
- POST
/api/v1/stt/post/Short-audio transcription - GET
/api/v1/stt/get/?page=1&limit=10v1 history list - POST
/api/v2/stt/post/Long-audio transcription (async) - GET
/api/v2/stt/get/v2 history list
Quick start #
1. Create an API key
Generate an API key in Space and send it in the
X-Api-Keyheader for every request.2. Send a test request
Validate the TTS/STT endpoints through cURL or backend code and confirm the response shape you expect.
3. Wire status and history
For async flows, implement
statusandhistoryto cover the full workflow.
Use X-Api-Key for server-to-server traffic. For large files and async tasks, implement history and status endpoints as part of the integration.
Machine-readable docs
A plain-text version of these docs is available for AI agents and LLMs.