Audio Separation

Audio Separation is an asynchronous API.

Create a task

Submit a multipart file with POST /v1/audio/separations.

import { readFile } from "node:fs/promises";

const form = new FormData();
form.append(
  "file",
  new Blob([await readFile("scene.wav")], { type: "audio/wav" }),
  "scene.wav",
);

const response = await fetch("https://taskinfer.com/v1/audio/separations", {
  method: "POST",
  headers: { Authorization: `Bearer ${process.env.TASKINFER_API_KEY}` },
  body: form,
});

if (!response.ok) throw new Error(await response.text());
const task = await response.json();
console.log(task.id, task.status);

Task lifecycle

Poll GET /v1/audio/separations/{id} until the task is completed or failed. The create response uses one of four states:

queued
in_progress
completed
failed

Poll the task

const response = await fetch(
  `https://taskinfer.com/v1/audio/separations/${task.id}`,
  { headers: { Authorization: `Bearer ${process.env.TASKINFER_API_KEY}` } },
);

if (!response.ok) throw new Error(await response.text());
console.log(await response.json());

Completed output

A completed task contains exactly three output URLs:

{
  "id": "631474364805546810",
  "object": "audio.separation",
  "status": "completed",
  "duration": 61.2,
  "created_at": 1786089600,
  "completed_at": 1786089642,
  "error": null,
  "output": {
    "dialogue": "https://...",
    "music": "https://...",
    "effects": "https://..."
  }
}

Output URLs are supplied by the processing provider and may be temporary. Download required results promptly.

Limits and billing

The endpoint accepts MP3, MP4/M4A, WAV, and WebM files up to 100 MB. It is billed at 20,000 credits per started audio minute (approximately $0.02 per minute).

Errors

For common HTTP failures, see Errors and troubleshooting.

Prepare authentication in Get started. For endpoint and request definitions, see the API Reference.