Voice boost for video — music and effects held back
An online tool for video where the speech is hard to make out: it splits the soundtrack into voices and background, lifts the speech and holds down music and effects. You hear the difference immediately — processing runs during playback rather than after a long conversion. The result can simply be watched, or saved as a separate file: the picture is copied untouched and only the audio changes.
Drop the file into the player — playback starts at once, with no conversion step to wait through.
Off, Light, Medium or Strong. Medium is the default; voice and background have their own sliders under “Settings”.
Hold the “Original” button (or the O key) to hear the untouched audio, release to come back.
Watch right here, or press “Save” — you get the same file with the sound you just dialled in.
Lift the voices above the music and keep the result
The file opens straight in the browser through Web Audio: playback starts at once and nothing is converted. Saving the fixed video is a separate step, on a button.
The audio track can be converted right here — the video itself is untouched. A two-hour film takes a few minutes, once.
Settings
Space — pause · ←→ — ±10 s · B — speech level · O (hold) — original · C — subtitles · T — wide screen · F — fullscreen
Save with the fixed audio
Keep this tab open. Speed depends on the length of the video and on your device.
Neural model: work out this moment
The normal mode computes the centre of the stereo image and answers instantly, but it has a limit: bass and drums sitting dead centre cannot be told from speech. The model separates for real — that costs a wait, so it works on a fragment around where you are rather than on the whole film.
Keep this tab open. Speed depends on your device.
