How much does the AI voice changer cost?

Trying the first 15 seconds costs 1 credit a voice, plus 2 for the try with the background kept. A whole recording costs 4 credits a minute, or 10 credits a minute with the background kept, and each started minute counts as a full one: a 3-minute voice-over is 12 credits. Every price is on the button before you start, and a run that doesn’t finish gives the credits back. See pricing for credit packs and plans.

Is there a free voice changer?

Pitch and speed effects run in your browser at no cost in the pitch changer. The AI voice changer, which makes your recording a different person, uses credits: 1 credit tries a voice on your first 15 seconds, and your sign-up credits change about 15 minutes.

Is it a real-time voice changer?

No: it changes recordings you already have, and videos. Record or upload the file, and a short recording is ready in a minute or two.

Can it change a male voice to a female voice?

Yes, pick a female voice: your timing and feeling stay, and the sound comes from her voice. Female to male works the same way. Between the sexes the result depends most on the recording, and some voices keep a trace of the original speaker, so try two or three on your first 15 seconds and keep the one that sounds right. A clear recording of one speaker, close to the microphone, works best.

Can I use my own cloned or designed voice?

Yes. Voices you designed from a description or cloned from your own sample are in the picker under My voices, next to the library.

What is speech to speech?

Speech to speech is the technique behind this AI voice changer: speech goes in, and the same speech comes out in another voice, with its words, rhythm and emotion kept. Text to speech, by contrast, reads written text; for that, use text to speech.

Does it keep background music?

Yes, when you turn on Keep the background: the voice is separated from the music and the room sound, changed, and mixed back where it was. With it off, you get the new voice on its own.

How long can a recording be?

Up to 30 minutes. Long recordings are cut at the pauses between sentences, changed in parts at the same time, and joined back exactly where each part came from.

Does it work in other languages?

Yes. The voice changes whatever language you speak, and the new voice says your words as you said them. Two of our samples are in Spanish and Chinese.

Is using an AI voice changer legal?

Changing your own recordings into the library’s voices, or into a voice you designed or cloned from your own sample, is what this tool is made for. Speaking as a real person without their permission is not allowed, and some places ask you to say when a voice is made with AI.

Can an AI voice be detected?

The MP3 and WAV files carry a label in their metadata saying they were made with AI. If you post a changed voice, saying so is good practice, and some platforms and places ask for it.

Are my recordings private?

Yes. Your recordings and their changed versions are visible only to you and stay in your library for 30 days.

AI Voice Changer for Recordings and Videos

plus the ones you design or clone: 400+ voices
per recording, cut at its pauses: 30 min
in up to 3 voices, side by side: First 15 s
and the room sound, if you like: Music kept
same picture, new voice: Video kept
a minute of recording: 4 credits

How do you change your voice in a recording?

To change your voice in a recording, upload the file or record yourself, pick the voice you want, and try the first 15 seconds. The AI voice changer says the same words with the same timing and emotion in the new voice. Then change the whole recording and download an MP3, a WAV or your video with its picture.

How to use the AI voice changer

Three steps, with a short try in a few voices before the whole recording.

Upload or record
Drop in an MP3, WAV, M4A or FLAC file, or a video such as MP4 or MOV, or record yourself in the studio. From a video only the sound goes up, taken out on your device. Recordings run up to 30 minutes.
Pick voices and try them
Choose up to 3 voices from the library, or your own designed and cloned voices, and hear the first 15 seconds in each, 1 credit a voice (2 more with the background kept). Switch between them and the original while it plays.
Change the whole recording
Keep the voice you like, decide whether the background stays, and change the whole recording. Download an MP3 or a WAV, or your video with the picture copied as it is.

What this AI voice changer does for you

A voice changer for recording, not for live calls: the new voice performs exactly the way you did.

A white theatre mask on dark green velvet

Your performance, in another voice

The voice changer listens to how you said it and says it again in the new voice: the same words, the same pace and pauses, the same whisper, laugh or shout. The changed recording runs exactly as long as yours, so it still fits your video or your edit.

400+ voices, and the ones you make

Narrators, characters, young and old voices in 36 languages, each with a clean clip prepared for changing into. A voice you designed from a description or cloned from your own sample works too, so a character can come back in every episode.

Try 3 voices on the first 15 seconds

Not sure which voice fits? Try up to 3 at once on the start of your own recording, 1 credit each, and switch between them at the same moment. Then change the whole recording in the one you like.

Keep the background music and room sound

Turn on Keep the background and the voice is lifted out of the recording, changed, and put back over the music and the room sound exactly where it was. A vlog keeps its soundtrack; a scene keeps its street.

Change the voice in a video, keep the picture

Drop in an MP4, MOV, MKV or WEBM file. Only its sound goes up, and the new voice goes back into the same video with the picture copied frame for frame, ready to post (a browser that can’t write the video’s sound gives you the new sound as MP3 or WAV).

Long recordings in one go

Recordings up to 30 minutes are changed whole: cut at the pauses between sentences into short parts, changed several at a time, and joined back exactly where each part came from, so a full podcast episode or an audiobook chapter goes in as one file.

AI voice changer samples: before and after

Each original is a public-domain recording or a line read by one of our voices, changed into another voice by this voice changer exactly as it came out. Switch between the original and the new voice while it plays.

AI voice changer guide: get the voice you want

Change your voice in a recording, step by step

Start with the recording you already have: a voice-over, a podcast segment, a line for a character. The voice changer keeps your words and your delivery, so perform it the way you want it to sound, then choose who says it.

1Upload the recording or the video, or record yourself on the page.
2Pick up to 3 voices and try the first 15 seconds.
3Switch between the voices and the original at the same moment.
4Change the whole recording in the voice you like, and download.

Male to female voice changer, and female to male

To change a man’s voice into a woman’s, or the other way round, pick a voice of the other sex. The new voice takes its sound from that voice, while the timing and the feeling stay yours.

Between the sexes the result depends most on your recording, and the voices differ more from one another than anywhere else. Some keep a trace of the original speaker, so try two or three before you choose: the 15-second try plays them side by side on your own recording. A clear recording close to the microphone, with one person speaking, works best.

Want the same voice only higher or deeper? The pitch changer moves its pitch on your device, without changing who is speaking.

Record well for a voice changer

Everything you do carries over: a pause, a breath before a line, a smile in the voice. Record in a quiet room, a hand’s width or two from the microphone, and say it the way the character would.

One speaker at a time works best. For a conversation, change each speaker’s part on its own and put them together in an editor, or keep one voice as it is.

When to keep the background

Keep the background for a recording with music or a place in it: a vlog with its soundtrack, a street interview, a scene with rain behind it. The voice is separated, changed and mixed back exactly where it was, at 10 credits a minute.

Leave it off for a voice recorded on its own, such as a voice-over or a podcast in a quiet room: you get the new voice clean, ready to put over any music, at 4 credits a minute.

AI voice changer or voice effect?

A voice effect, such as a deep voice, a robot or a chipmunk, bends the same voice: it is still you, only lower, higher or filtered. The pitch changer and the reverse audio tool do that on your device.

An AI voice changer makes it a different person: another timbre, age and accent, as if someone else had performed your recording. It is also called speech to speech, because speech goes in and the same speech comes out in another voice.

Voices for narration, characters and audio drama

Narration: a warm or a deep narrator for a video essay, a documentary voice for a product film, a calm reader for an audiobook chapter you recorded yourself.

Characters: an old sea captain, a young hero, a gentle grandmother, all performed by you and voiced by the library. Design a voice of your own for a character who returns, and it is there for every episode.