Can I use this deep voice changer live on Discord or a call?

No. It works offline on a file or a recording you make on the page, and you download the result. It cannot sit between your microphone and Discord, a game or a phone call.

How deep does the voice go?

Between 4 and 6 semitones down, with 2 to 5 dB of added bass, and the slider moves both. That multiplies every frequency by about 0.79 to 0.71. For other amounts, use the Audio Pitch Changer, which has no bass shelf.

Does it make the recording slower or longer?

No. A deep voice converter that only slowed the audio would drag your words out. This one changes the pitch and keeps the length, so the file ends at the same moment as the original.

Will it make me sound like a different person?

Not reliably. Your words, pace and accent stay the same, so anyone who knows you may still recognize you. It is a sound effect, not a privacy tool. Use it on recordings you have the right to change, and do not pass the result off as someone else.

Does my audio leave my device?

No. The pitch shift and the new file are both made in your browser. The page counts ordinary analytics events, such as the slider range and the file format, but never the sound or the file name.

Deep Voice Changer

Get a deep voice from a file or a recording: this page lowers the pitch 4 to 6 semitones, adds a little bass and keeps the length the same. It works offline on saved audio, so it is not a live changer for calls or Discord. Processing happens in your browser, so the file is not uploaded.

Start from

Drop a recording here or click to browse

MP3, M4A, AAC, WAV, OGG, OPUS, FLAC, WEBM or MP4, up to 100 MB and 30 minutes.

Your recording is processed in your browser and is not uploaded.

How to make your voice deeper

The Deep voice is already selected on this page, so you only choose how far down it goes.

1
Load your voice
Drop an MP3, M4A, WAV or other audio file up to 100 MB and 30 minutes, or record up to 10 minutes with your microphone. One person speaking works best.
2
Choose how low
The slider runs from 4 semitones down at the left to 6 at the right. Listen to the first 30 seconds as Original and Changed, and try a sentence you know well.
3
Save the deep voice
Press Change the voice to convert the whole recording, then download it as an MP3 or a 16-bit WAV. Longer files take a while, because the pitch shift runs on your device.

Part of the Voice Changer, which has seven other voices. Want a shift smaller than 4 semitones or larger than 6? The Audio Pitch Changer goes up to 12 either way. For a buzzing machine voice, try the Robot Voice Generator.

What a pitch shift does to a voice

A semitone is one step on a piano. Going down one multiplies every frequency by about 0.944, and twelve of them make an octave, which halves every frequency. The formula is 2 to the power of (semitones ÷ 12), so 5 semitones down multiplies everything by 0.749.

The pitch you hear is the voice's fundamental frequency, set by how fast the vocal folds vibrate. Typical adult ranges are about 90 to 155 Hz for men and 165 to 255 Hz for women. So a voice at 120 Hz lands near 90 Hz at the middle of the slider, and one at 210 Hz lands near 157 Hz.

This kind of shift also moves the formants, the resonances of the throat and mouth, by the same factor. That is part of why it sounds big and not only low: in a 2005 study of scaled vowels, vocal-tract length had a strong influence on how large listeners judged a speaker to be. A real deep-voiced person, though, did not get there by having every resonance and every breath moved down by an exact ratio, and that difference can be audible.

How low each slider position goes

Two example voices, 120 Hz and 210 Hz. Measure your own with the Voice Pitch Analyzer if you are curious.

Slider	Semitones	Frequencies	120 Hz voice	210 Hz voice	Bass added
0	-4	× 0.794	95 Hz	167 Hz	+2 dB
25	-4.5	× 0.771	93 Hz	162 Hz	+3 dB
50	-5	× 0.749	90 Hz	157 Hz	+3.5 dB
75	-5.5	× 0.728	87 Hz	153 Hz	+4.5 dB
100	-6	× 0.707	85 Hz	148 Hz	+5 dB
One octave (not on this page)	-12	× 0.500	60 Hz	105 Hz	None

From this page's deep preset. Bass is a low shelf at 200 Hz. The slider starts at 50.

How many semitones sound natural?

There is no agreed number. The sources read for this page give no threshold, and it depends on the voice you start with and on who is listening. What can be measured is where the shifted pitch lands against the typical ranges above.

Take the 120 Hz voice in the table. At 4 semitones it sits at 95 Hz, inside the typical male range. At 5 it is right at the bottom, and at 6 it is at 85 Hz, just below. That is a fair place to start if you want the result to pass as a person. The 210 Hz voice leaves the typical female range at about 4.5 semitones and reaches the typical male range at about 5.5.

A full octave is a different thing: the 120 Hz voice drops to 60 Hz, far under the typical range, and it usually sounds more like a monster or a slowed-down tape than a person. This page stops at 6. If you want the cartoon version on purpose, the Audio Pitch Changer gets you there.

Why the result can sound processed

A shifted voice rarely passes as a real one at every setting. These are the usual reasons, and none of them is a fault in your recording.

The resonances move too
This page shifts everything together and does not try to hold the formants in place. Tools that do keep them analyze the voice and rebuild it. Scaling them, as here, is what Wikipedia's article on pitch scaling describes as distorting the formants.
The audio is cut into small pieces
To keep the length, the recording is squeezed in overlapping pieces of 40 milliseconds and then read back slower. Each piece is lined up with the last by matching the wave, which suits one clear voice. Music, room echo or two people talking have less to line up on, so they can flutter.
The top end falls
An s around 8,000 Hz drops to about 5,700 Hz at 6 semitones, so consonants get duller and the voice can sound muffled. Nothing replaces the frequencies that moved down.
The bass boost lifts noise as well
The low shelf adds 2 to 5 dB around and below 200 Hz. It also lifts room hum, desk thumps and the pop of a p, all of which are now lower and easier to notice.

Getting a more believable deep voice

Start from a clean recording
Quiet room, one speaker, close to the microphone. Noise and echo get shifted along with you, and the shelf makes them louder.
Start at the left of the slider
Try 0 first and go up only if you need to. Your words, pace and accent do not change at any setting, so you may still recognize yourself.
Speak a little slower
A shift keeps your timing. Fast speech at a low pitch can sound odd, so take a slightly slower pace when you record and leave more room around your words.
Want bass without the pitch?
If your voice only sounds thin, the Bass Booster adds low end and leaves the pitch alone. That is the cleanest option, since nothing is cut up.

Related: Bass Booster and Voice Pitch Analyzer.

Need the words from a recording, not a new voice?

This page changes how a recording sounds. ParrotNotes does a different job: it records calls and meetings on your phone, transcribes them and summarizes them.

Get Started Free

Related tools