API Documentation - Tích Hợp Text to Speech

All tasks in the History section now have a Report Bug button (flag icon). If your task runs into an error, please let us know so we can fix it and give you a better experience. Thank you!

API Key

KingCong Public API

One API key, one base URL, all supported speech providers.

https://api.kingcongstudio.com

ElevenLabsexpressive multilingual voices, model controls, transcript export

MiniMaxfast speech models through KingCong AI33/AI84 routing

KingCongnative KingCong voices, SRT input/export, Vietnamese-focused models

Edge TTSlow-cost Microsoft style multilingual voices

Kokoroopen-source style low-cost voices

Vbee / Fish AudioAI33 library providers where available

Authentication

Use your API Docs key, not the tool login key. New keys use the kingcongapikey_ prefix. Old API Docs keys still work.

Health Check

curl "https://api.kingcongstudio.com/v1/health-check" \
  -H "xi-api-key: $API_KEY"

Credits

curl "https://api.kingcongstudio.com/v1/credits" \
  -H "xi-api-key: $API_KEY"

Providers, Models, Voices

List supported providers, model IDs and voice IDs before creating a task.

GET /v1/providersProvider metadata and supported features.

GET /v1/models?provider=kingcongModels for one provider. Omit provider to get all.

GET /v1/voices?provider=kingcong&search=team&limit=20Paginated voice list with search.

curl "https://api.kingcongstudio.com/v1/voices?provider=kingcong&limit=20" \
  -H "xi-api-key: $API_KEY"

Provider IDs

elevenlabsElevenLabs voices and models.

minimaxMiniMax voices and speech models.

kingcongKingCong native voices and models.

edge_ttsEdge TTS voices.

kokoroKokoro voices.

Create Speech

Create a TTS task using any supported provider. Poll the returned task_id until audio_url is ready.

curl -X POST "https://api.kingcongstudio.com/v1/text-to-speech/$voice_id" \
  -H "Content-Type: application/json" \
  -H "xi-api-key: $API_KEY" \
  -d '{
    "text": "Xin chào, đây là đoạn kiểm tra KingCong Public API.",
    "provider": "kingcong",
    "model_id": "vieten_speech",
    "voice_name": "CD Team 2",
    "speed": 1,
    "vol": 1,
    "pitch": 1,
    "with_transcript": false
  }'

textREQUIREDInput text. KingCong requires at least 100 characters for normal text.

providerREQUIREDUse elevenlabs, minimax, kingcong, edge_tts or kokoro.

model_idUse a model from /v1/models. Default is provider-specific.

voice_nameOptional display name stored with the task.

with_transcriptOptional SRT/JSON transcript export when supported.

export_srtKingCong-specific SRT export flag.

Task Management

Read task status, list history, delete a task, or check credits.

curl "https://api.kingcongstudio.com/v1/task/$task_id" \
  -H "xi-api-key: $API_KEY"

curl "https://api.kingcongstudio.com/v1/tasks?page=1&limit=20" \
  -H "xi-api-key: $API_KEY"

curl -X DELETE "https://api.kingcongstudio.com/v1/task/delete" \
  -H "Content-Type: application/json" \
  -H "xi-api-key: $API_KEY" \
  -d '{"task_id":"uuid_task_id"}'

AI Markdown Guide

A compact prompt-friendly spec you can paste into an AI agent before asking it to call the API.

# KingCong Public API

Base URL: https://api.kingcongstudio.com
Auth header: xi-api-key: $API_KEY
API Docs keys are separate from tool login keys.

## Find available voices and models
- GET /v1/providers
- GET /v1/models?provider=kingcong
- GET /v1/voices?provider=kingcong&search=team&limit=20

## Create speech
POST /v1/text-to-speech/{voice_id}

Body:
- text: required string
- provider: elevenlabs | minimax | kingcong | edge_tts | kokoro
- model_id: model from /v1/models
- voice_name: optional display label
- with_transcript: optional boolean
- export_srt: optional boolean for KingCong
- speed, vol, pitch: optional KingCong controls

## Notes
- KingCong text requires at least 100 characters.
- Use /v1/task/{task_id} to poll result.
- Done tasks include audio_url and optional srt_url/json_url.