SDK & API Reference
Build with the
Decrackle Audio AI
Platform
Transcribe speech, remove noise, and manage audio — all through a single TypeScript SDK or REST API.
npm install @decrackle/sdkEverything you need
🎙
Speech to Text
Batch and streaming transcription with auto language detection across 30+ languages.
ASR docs →🔇
Audio Denoising
AI-powered noise removal using EnhanceNet models. Upload any audio, get back studio-clean output.
Denoising docs →⚡
TypeScript-first SDK
Full types, zero runtime dependencies. Works in Node.js 18+ and modern browsers.
Installation →🔁
Production-ready
Automatic retries, configurable timeouts, cancellation support, and typed error classes.
Error handling →🔑
API Key Auth
Scoped API keys with plan-based access control. Set once via env var, works everywhere.
Authentication →💾
Storage Management
Presigned URLs for secure file access. Track quota usage and manage jobs programmatically.
Storage docs →Up and running in minutes
import { Decrackle } from '@decrackle/sdk';
const client = new Decrackle({ apiKey: process.env.DECRACKLE_API_KEY });
// Transcribe audio
const result = await client.asr.transcribe(audioBuffer, { language: 'en' });
console.log(result.transcriptionText);
// Denoise audio
const clean = await client.denoise.process(audioBuffer);
console.log('Clean audio:', clean.outputUrl);