Speaker Split
Remove one speaker from a recording and keep everyone else
Upload an interview, podcast, meeting or phone call. SpeakerSplit works out who is speaking when. Pick the person you want gone, and download the whole recording, silent wherever they speak.
How to remove a speaker
- Open Speaker Split and upload your recording. Audio and video files both work.
- Wait a few minutes. Each speaker gets their own track.
- Play the tracks to find the person you want to take out. You can rename each speaker.
- Under the tracks, in "Remove a speaker", click that person. Click more people to take out several.
- Click "Download without". You get the whole recording, silent wherever they speak.
What you get
- One file as long as your recording, so it still lines up with your video.
- Everyone else’s voices, and any sound between the removed person’s turns.
- MP3, WAV, FLAC, AIFF, M4A, OGG or OPUS. Pick one before you download.
- Every speaker on their own track as well, if you only want one person.
Good to know
- Removing a speaker costs no extra credits. You pay once, for the split.
- While the removed person talks, that part is silent, background included.
- If they talk over someone else, that moment goes silent for both of them.
- The language is detected for you.
What it costs
| Recording length | Credits |
|---|---|
| Under 5 minutes | 1 |
| 5 to 30 minutes | 2 |
| 30 to 60 minutes | 4 |
| 60 to 90 minutes | 6 |
| Each started 30 minutes after that, up to 4 hours | +2 |
Removing a speaker after the split costs nothing extra. Plans and credit options are on the pricing page.
Common questions
How do I remove one person’s voice from a recording?
Upload it to Speaker Split. When the split is done, click that person under "Remove a speaker" and download. You get the whole recording, silent wherever they speak.
Does removing a speaker cost extra?
No. You pay credits once to split the recording. Removing one or more speakers after that is free.
What happens when two people talk at the same time?
That moment goes silent for both, because the cut follows who is speaking when. It works best when people take turns.
Can I keep only one speaker instead?
Yes. Every speaker already has their own track. Download the one you want.