Saturalabs / Voice & video
Isolate voice from video. Keep the conversation.
Pull spoken dialogue out of a video’s soundtrack. Describe the voice you want, audition the separated audio, and bring the result back into your video editor.
Start with your recordingThe track you’re looking for
A separated dialogue track for your video editing workflow.
Name one audible source. You can edit this prompt before running the separation.
Uploading opens the editor. Downloads require an account and credits. View pricing.
A focused approach
Extract the voice, then finish the video in your editor
A video soundtrack often arrives as one finished mix: dialogue, a music bed, and environmental sound together. If you no longer have the original microphone track, voice isolation gives you a way to try separating the audible dialogue from that mix. Upload the video directly; you do not need to extract an audio file first.
Saturalabs returns separated audio, rather than a newly rendered video. Keep Isolated sound for dialogue, then import the WAV into your video editor. Align it with the original clip, mute or lower the old audio, and check sync before exporting. Preserve an untouched copy of the original video throughout the edit.
From mix to source
How to isolate voice from video
Prepare a manageable clip
Use an MP4 or MOV file up to 50 MiB (52.43 MB). For a larger project, export the relevant scene in your video editor. Note its original start point so you can align the replacement audio later.
Request the dialogue
Try ‘The foreground spoken dialogue’ for a scene with a clear speaker. Use ‘The narrator’s voice’ for voiceover. Request speech rather than a list of sounds to remove.
Check both audio tracks
The voice belongs in Isolated sound. Listen to difficult words and transitions, then check Background audio for speech that may have been left behind.
Put the audio back in the timeline
Download the dialogue WAV and place it at the same clip start in your editor. Check lip sync near the beginning and end, adjust levels, and export the video from that editor.
Be specific about the sound
A prompt for the result you want
Describe the source, then choose the output that matches your goal. These are starting points to try with your own recording.
| Your goal | Try this prompt | Keep this track |
|---|---|---|
| Recover an on-camera speakerUse for speech you can already hear clearly enough to identify. | The foreground spoken dialogue | Isolated sound |
| Extract a voiceoverA more specific target when narration sits above music and effects. | The narrator’s voice | Isolated sound |
| Keep an interview conversationThis requests the conversation as a group, rather than promising separate tracks per speaker. | The interview participants speaking | Isolated sound |
Made for a real next step
Put the separated audio to work
Talking-head videos
Try recovering a presenter’s speech when music or ambient sound competes with the message. Review every important sentence before publishing.
Documentary interviews
Separate dialogue from a location recording for an edit. You may want to retain some original ambience underneath so cuts do not feel abruptly silent.
Repurposed clips
Prepare a dialogue track for a shorter social edit or a new mix when the source recording is available but separate audio stems are not.
For the finished video, follow the audio replacement and sync guide to place the chosen WAV, mute the source sound, and check the final export.
Listen before you commit
Three checks worth making
Compare a quiet passage and a busy passage with the original, at a similar listening level.
Dialogue continuity
Listen through pauses, turn-taking, and scene cuts. Check whether quieter speakers disappear or background speech returns.
Lip sync
Use the same timeline position as the source clip. Confirm that mouth movements and words still align after importing the WAV.
The final mix
Audition the voice with your intended music and ambience. A solo track can reveal artifacts that sit differently in a complete video mix.
Know the limits
A useful dialogue track still needs an editorial check
Voice isolation is not a complete video post-production process. It does not add subtitles, re-edit the picture, replace the video's soundtrack automatically, or fix words hidden by severe distortion. The result is audio for you to review and use in your editor.
Music and voices can overlap in ways that leave traces or affect the extracted speech. Strong echo, crowd chatter, and several simultaneous speakers can make the task harder. If the original production sound is available, compare it before choosing an isolated version.
Can I extract a voice from an MP4 without converting it first?
Yes. Saturalabs accepts MP4 and MOV video directly, as well as MP3 and WAV audio. The uploaded file must be 50 MiB (52.43 MB) or smaller.
Will I get a video with the voice already replaced?
No. The separated outputs are audio tracks. Download the dialogue WAV and combine it with your original video in your video editor.
Which output contains the extracted dialogue?
When your prompt requests spoken dialogue, Isolated sound contains that target. Background audio contains the remaining soundtrack. Listen to both to check the separation.
Can I separate music from dialogue instead?
Yes, you can try targeting the background music and keeping Background audio for the rest of the soundtrack. That workflow is useful when you also want to retain environmental sounds, but you must check for lost dialogue and music remnants.
How do I avoid audio sync problems?
Place the downloaded WAV at the same start point as the clip you uploaded. If you exported only part of a longer video, use that segment's timeline position. Check the beginning, middle, and end before the final export.
Start with a sound you know
Bring your own recording.
Describe one source. Listen to both tracks. Keep what works for your project.