Server-side processing with htdemucs. Separate vocals, drums, bass, guitar, piano and more from any audio file.
Upload Audio
🎵Drop your audio file here
or click to browse — MP3, WAV, FLAC, OGG, AAC, M4A…
⚠️ File exceeds the 20-second limit
Free accounts can process up to 20 seconds. Please split your file with our audio splitter, or buy tokens for unlimited duration and advanced options.
⭐ WAV output requires a Premium account
WAV (lossless) export is available to Premium subscribers. Upgrade your account or choose MP3 or FLAC.
Settings
Settings greatly affect processing time. Expect up to 1h of processing per minute for maximum quality. Once the job is done it gets transferred to your device.
🎵 2 stems🤖 htdemucs⚡ shifts = 1📅 auto segment
Number of Stems
Vocals & accompaniment only. Fast and focused.
Hybrid Transformer. Best balance of speed and quality.
1
Each shift runs separation at a different time offset and blends results. Higher = better quality, slower. 1 (fast) to 5 (highest quality).
Shorter segments use less memory but may introduce more boundary artefacts. HT models cap at 7s (trained on 7.8s segments); MDX models use fixed internal segmentation.
Higher overlap reduces boundary artefacts at the cost of extra processing time.
Output Format
Bitrate
Lossless FLAC — no quality loss, larger file size.
Bit Depth
Sample Rate
⭐ Premium Processing
Processing threads
More threads finish faster. Cost is 10 tokens per audio-second per thread, multiplied by the separation quality (shifts) — so threads and quality both raise speed and price. If the shared pool is busy you'll be asked to start with fewer threads or wait.
When the job finishes
Save-to-storage needs enough free space and is kept until you delete it. Browser results stay available for 48 hours.
⏱ Est. time: –🎫 Est. cost: – tokensBalance: – tokens
Processing
queued
Position - of - in queue — your job will start shortly.
0%
Step:-ETA:-
Results
We use cookies to improve performance, analytics, and marketing.
Essential cookies are always active.
