Aisha AI
by voicelab

Aisha API hujjatlari

O‘zbek nutqi uchun REST API’lar - text-to-speech, speech-to-text va voice agentlar. API key bilan autentifikatsiya qilib, birinchi so‘rovingizni bir necha daqiqada yuboring.

API key olish Jamoa bilan bog‘lanish Base URL https://back.aisha.group

Dasturchilar uchun

SDK bilan tezkor start #

SDK + CLI

aisha-ai - Node.js (npm) va Python (pip) uchun dependency’siz SDK va CLI. O‘rnatib oling yoki npx bilan darhol ishga tushiring.

Node.js · npm

npm install aisha-ai

Python · pip

pip install aisha-ai

Bitta buyruq bilan

export AISHA_API_KEY=your_api_key
npx aisha-ai tts "Salom dunyo" --model Gulnoza --out salom.wav

AI agent bilan ishlayapsizmi? Unga llms-api.txt faylini bering - butun API bitta matn faylida, yoki aisha-ai paketini o‘rnatishni so‘rang.

llms-api.txt
npm’dagi aisha-ai PyPI’dagi aisha-ai GitHub’dagi manba

Bizni PyPI va NPM’da ham bemalol topa olasiz.

Tillar bo‘yicha SDK’lar

Mobil va desktop

Ovoz modellarimizni har kuni yaxshilab bormoqdamiz.

TTS API #

API v1 TTS Sync / Async Realtime

Text-to-Speech endpointlari. Server integratsiya uchun har bir so'rovda X-Api-Key header yuboring.

  • Audio yaratish uchun POST /api/v1/tts/post/ endpointiga transcript yuboring.
  • webhook_notification_url yuborilsa, so'rov async ishlaydi va 202 Accepted qaytadi.
  • Natijani GET /api/v1/tts/status/{id}/ yoki history orqali tekshiring.
  • Realtime TTS uchun wss://back.aisha.group/api/v1/tts/realtime WebSocket yoki back.aisha.group:443 gRPC ishlating.
  • Bitta connection ichida bir nechta text turn yuboring; har turn uchun qayta ulanishingiz shart emas.
  • TTS character tarifi stream boshlanishida balance tekshirilgandan keyin qo‘llanadi.

Endpointlar

  • POST /api/v1/tts/post/ Audio yaratish
  • GET /api/v1/tts/status/{id}/ Status tekshirish
  • GET /api/v1/tts/get/?page=1&limit=10 History ro'yxati
Reference ochish

STT API #

API v1 API v2 Realtime STT

Speech-to-Text endpointlari. Server integratsiya uchun har bir so'rovda X-Api-Key header yuboring.

  • Qisqa audio uchun POST /api/v1/stt/post/ (sync).
  • Uzun audio uchun POST /api/v2/stt/post/ (async, task_id qaytadi).
  • Realtime audio uchun wss://back.aisha.group/api/v1/stt/realtime WebSocket ishlating.
  • WebSocket bitta connection ichida davomli session uchun ishlatiladi; tugatish uchun {"event":"end"} yuboring.
  • Server streaming integratsiyasi uchun back.aisha.group:443 gRPC endpoint mavjud, lekin har Transcribe chaqiruvi bitta audio oqimi uchun.

Endpointlar

  • POST /api/v1/stt/post/ Qisqa audio transkripsiya
  • GET /api/v1/stt/get/?page=1&limit=10 v1 history ro'yxati
  • POST /api/v2/stt/post/ Uzun audio async transkripsiya
  • WS /api/v1/stt/realtime?format=webm&token=YOUR_API_KEY Realtime WebSocket transkripsiya
Reference ochish

Qisqa qo‘llanma #

  1. 1. API key oling

    Space panel orqali API key yarating va so‘rovlaringizga X-Api-Key header sifatida qo‘shing.

  2. 2. Test request yuboring

    cURL yoki backend kodingiz orqali TTS/STT endpointlarga test so‘rov yuborib, response formatini tekshiring.

  3. 3. Status va history ni ulang

    Async oqimlarda status va history endpointlari bilan flow ni yopib chiqing.

Server integratsiyasi uchun X-Api-Key ishlating. Katta fayllar va async vazifalar uchun history va status endpointlarini ham implement qiling.

Mashina o‘qiydigan hujjatlar

AI agentlar va LLM’lar uchun hujjatlarning matnli versiyasi ham mavjud.

AI agentlar uchun llms-api.txt