20
🔍

Separate vocals and instrumentals

Upload audio, record from your microphone, or paste a supported public YouTube, X, Instagram, or Facebook video link. Download the isolated vocal and instrumental tracks when processing is complete.

PLATFORM FACTS
$6
starting credit pack
100+
languages
6
creator tools
Flexible
monthly or one-time
🔍 VOCALS🔍 STEMSECONOMY: 20 🎟️ = 2.5 minECO: 20 🎟️ = 2.5 min
🧳
Check in your file
Upload / drag & drop or Microphone recording is not available in this browser. You can still upload an audio file.

Protected by reCAPTCHA — Privacy & Terms.

HOW IT WORKS

How It Works

01

Upload or Paste URL

Upload an audio file, record from your microphone, or paste a supported public YouTube, X, Instagram, or Facebook video link. We handle MP3, WAV, FLAC, and most formats.

02

AI Separates the Stems

SpeakSwap analyzes the mix and produces separate vocal and instrumental files. Separation quality varies with the recording, arrangement, and effects.

03

Download Your Stems

Get the isolated vocals, the clean instrumental, or both. Perfect for remixes, karaoke, or sampling.

30-SECOND CC0 TEST

Hear a real vocal-remover result

This is the same excerpt before and after SpeakSwap separated it into vocal and instrumental tracks on August 10, 2026. Listening is free and does not start a job.

Original mix
Vocal output
Instrumental output

What to listen for

Compare the original with the instrumental during sung phrases. The instrumental reduces the lead voice while retaining most of the accompaniment. The vocal file keeps the lead performance, although faint instruments and artifacts can remain where sounds overlap.

How this test was made

We processed seconds 60 to 90 of the official MP3 through the production SpeakSwap vocal-remover pipeline. No cleanup or mastering was applied after separation. All three files are 192 kbps MP3s.

Source: Ua Like No A Like by Rita Braga. Free Music Archive lists the track under CC0 1.0 Universal. At its measured 30.04-second duration, a run uses 5 credits under whole-credit rounding at the current 8-credit-per-minute rate. Your dollar cost depends on the credit pack. Only process audio you own or have permission to use.

FARE CHECK

Compare options

ProviderPriceModel
SpeakSwap$0.06/min on Scale planPay as you go or subscription
LALAL.AI Pro$0.07/min$35 for 500 min
StemSplit$0.10/minPay-as-you-go
Moises$4-14/moSubscription

Best rate

FAQ

Frequently asked

We support MP3, WAV, M4A, FLAC, OGG, and most common audio formats. You can also paste a supported public YouTube, X, Instagram, or Facebook video link and we'll extract the audio.
zZ