What each effect does to the sound
Every effect is a classic audio process, run by the audio engine built into your browser. In plain words:
| Effect | What it does |
|---|---|
| Deeper | Lowers the pitch by 4 semitones and lifts the low frequencies a little, for a warmer, deeper sound. |
| Higher | Raises the pitch by 4 semitones and keeps the speed the same. |
| Chipmunk | Raises the pitch by 9 semitones and trims the deep bass, so the voice sounds small and thin. |
| Monster | Lowers the pitch by 8 semitones, boosts the bass, adds a little distortion and a short room echo. |
| Robot | A ring modulator: the voice is multiplied by a steady 50 Hz tone, with a very short metallic echo. |
| Radio | Cuts everything below about 400 Hz and above about 3,200 Hz, like a phone line or a small speaker, with light distortion. |
| Echo | Repeats the voice every 0.28 seconds, each repeat quieter and duller than the last. |
| Hall | Reverb from a generated 2.8 second room response, so the voice sounds like it is in a large space. |
- The pitch slider moves any effect up or down by up to 12 semitones. 12 semitones is one octave.
- With keep the original speed on, the pitch changes and the length does not: the audio is stretched in short overlapping slices, then played back faster or slower. Very large shifts can sound a little artificial.
- With it off, pitch and speed change together, like a tape played faster or slower: higher is shorter, deeper is longer.
Voice effect, AI voice or voice clone?
A voice changer like this one keeps your words and your delivery and changes how they sound. An AI voice or a cloned voice speaks new words from a script. Which you need depends on the job:
| This voice changer | Percify | |
|---|---|---|
| What it does | Adds an effect to your recording | Speaks a script in an AI voice, or in your own cloned voice |
| Is it still you? | Yes, your own voice with an effect on it | An AI voice, or a clone of your voice reading new words |
| What you give it | A recording or an audio or video file | A script, plus a short sample to clone a voice |
| What you get | A WAV file | An audio file, or a talking photo at 2 credits a second |
| Cost | Free, no account | Cloned voice: 2 credits for up to 40 words, then 1 credit per 15 words. Plans start at $12 a month (Starter, 450 credits) |
| Where it runs | In your browser, nothing uploaded | In the Percify app |
- Creating a Percify account needs no card; making voices and videos needs a plan.
Recording tips
- Record in a quiet room. Effects like robot, monster and hall make background noise louder and stranger too.
- Keep the phone or mic about a hand’s width from your mouth so loud words do not distort.
- Speak a little slower than usual for chipmunk and higher: raised voices are harder to follow.
- Leave a second of silence at the end so echo and hall can ring out.
- Use your own voice, or a recording you have permission to change and share.
Checked on 18 October 2026 against the Web Audio API in current browsers and the published prices on percify.io/pricing.