Vocal Separation
Standalone Vocal Separation splits video or audio into tracks you can preview and download separately. It is different from internal separation used during translation; translation projects do not automatically expose all standalone stems.
Vocal Separation
Standalone Vocal Separation splits video or audio into tracks you can preview and download separately. It is different from internal separation used during translation; translation projects do not automatically expose all standalone stems.
Choose a scene
| Scene | Delivered tracks | Format |
|---|---|---|
| General audio/video | Vocals and background, 2 tracks | MP3 |
| Music / songs | Vocals and accompaniment, 2 tracks | MP3 |
| Drama / film | Vocals, music, and sound effects, 3 tracks | MP3 |
| Speech / narration | Vocals, music, and sound effects, 3 tracks | MP3 |
Steps
- Open Vocal Separation and select the scene closest to the source.
- Upload one video or audio file and wait for the upload and duration check.
- Review the scene, estimate, and validation messages, then submit.
- Preview each completed track and download the MP3 files you need.
Sources and quality
The entry accepts MP4, MOV, WebM, AVI, MKV video and MP3, WAV, M4A, AAC, OGG, FLAC audio. Plan size, duration, and storage limits still apply. MP3 output does not mean the input must be MP3.
Echo, overlapping sources, and recording quality can affect separation. Outputs are stems, not translations, subtitles, or new speech. Use Audio/Video Transcription or Translation for translated subtitles.