Skip to content

40% off every plan

See plans
facecam

Free Trollge Voice Generator

Type the incident, hear it in the deep, distorted trollge narrator voice.

Updated · Tested by the Facecam team

Character Voice Generator: type it, hear a character

The trollge voice, before and after

  • Trollge chain

    Atlas, slowed, ten semitones down, with grit and a dark room

    Incident videos, creepy slideshows, the twist at the end of an edit

    “Incident forty-seven. The phone rang at three in the morning. Nobody was calling.”

  • Atlas, no effect

    Deep and measured American man, the voice under the effect

    Hearing what the chain starts from, or a calm narrator without the horror

    The same line from Atlas at 0.85× speed, dry.

How to make trollge text to speech in 3 steps

  1. Type the incident

    The tool above opens on the Trollge style with its own line. Replace it with yours, up to 5,000 characters.

  2. Generate

    Press Generate. The speech is made on our server; the deep, distorted chain is added in your browser.

  3. Tune it and download

    Too clean? Pick Pitch −2 or switch the effect to Monster. Re-rendering a line you already made doesn't cost anything. Then save the WAV.

What people make with the trollge voice

  • Incident videos

    The classic format: black-and-white images, a slow text crawl and the narrator reading each phase. Cut one sentence per image.

  • Creepy edits and shorts

    Put the deep read under a jump cut or a slow zoom for the last line of a horror short.

  • Lore and ARG channels

    Case files, logs and “recovered” messages read in a voice that sounds like it shouldn't exist.

  • A scare on stream or in Discord

    Play a prepared line from the Soundboard into voice chat at the worst possible moment.

    Discord Soundboard

What the trollge voice is (and isn't)

Incident videos are narrated flat and slow, pitched far down, with grit on the edges and a dark room around the voice. Here that is three steps: the deepest character voice, Atlas, at 0.85× speed; the Demon effect, ten semitones down with a bigger, darker tone; and extra distortion and a reverb tail. The settings table lists every value, so you can rebuild it in any editor.

It isn't the voice from any particular video. Trollge creators used whatever TTS and editing they had, so there's no single original; this is our own voice and effect chain set up for that sound.

The exact settings

SettingValue
VoiceAtlas (ElevenLabs stock voice: deep and casual, american)
Speed0.85×
EffectDemon, adjusted as below
Noise gatemutes the mic below -50 dB, between words
Pitch-10 (lower)
Tone-5 (sounds bigger)
Distortiondrive 0.45, 60% mix
Chorus / vibratowobbles 0.3 times a second, 30% mix
Bass+4 dB below 120 Hz
Reverb25% mix, 2.5 s long
TextUp to 5,000 characters per line

Clone your voice, then make it say anything

Record 5 to 30 seconds of your voice once. Then type any line and hear it in your voice, or narrate your videos with it.

Clone your voice

More about this voice

Write it like an incident

  • Open with the number: “Incident forty-seven.” Numbers read best spelled out.
  • Short sentences. The slow voice stretches every word.
  • End on the twist, then stop. Silence does the rest in the edit.

Trollge voice generator FAQ

What is a trollge voice generator?

A text-to-speech voice that reads your line slow, very deep and slightly distorted, like the narration in trollge incident videos. You type the incident and download it as a WAV.

Is it the voice from the original trollge videos?

No. Those were made with whatever TTS and editing each creator had. This is our own voice and effect chain set up for the same deep, slow, distorted sound.

Can I make it scarier?

Lower the pitch a step or two, or switch the effect to Monster or Cave Echo. Re-rendering the same line with new settings is free.

Can I use it in YouTube videos?

Yes. Download the WAV and use it in your own videos, monetized ones included.

Do I need an account?

No. Free to try: 3 lines a day, no account. Beyond that, a paid plan. The speech is made on our server; the effects run in your browser.