Docs & Help

Everything you need to sound great

Plain-English guides for setting up I Can't Sing, singing with live auto-tune, taking over any AI song, and fixing the little things. No jargon — if something's unclear, email support@icantsing.com.

Getting started

Three steps from download to singing. It takes about five minutes.

1. Download & install

Grab the Windows installer from the Download page— one file, about 500 MB, and the same download works for Free, Pro, Studio, and Lifetime. Double-click it and follow the prompts.

Seeing a Windows security warning? Stop and verify the publisher and digital signature from the official download first. If Windows cannot verify the file, contact support@icantsing.com rather than bypassing the warning.

2. Create your account & add your key

Open I Can't Sing and create an account with your email. If you've bought a plan, paste the license key from your email into Account. No key yet? Free mode works right away — you can start singing before you pay a cent.

The very first launch does a quick one-time setup (voice models + audio drivers). Give it up to a minute; after that it opens in a couple of seconds.

3. Allow your microphone

When Windows asks, allow microphone access. Then open Settings → Audio Devices and pick your mic as the input and your headphones as the output. Headphones matter — they stop the speakers from feeding back into your mic. Any mic works to start; a USB mic or audio interface sounds and feels better.

System Ready / First Sound

Before a rehearsal or show, use the System Ready control in the desktop app for an honest local preflight.

Choose devices

Select a real microphone/input and headphones, speakers, or interface output in Audio Devices. Virtual devices are flagged so you can verify their downstream route yourself.

Run the check

System Ready reads the already-available device list and current engine-health counters. It does not start audio, change routing, or touch your device settings.

Soundcheck on purpose

Start local audio only when you are ready, begin at a safe monitor level, and repeat the check to review reported device latency, DSP load, xruns, callback fallbacks, and recoveries.

What it cannot tell you yet: System Ready is not a microphone, room-feedback, clipping, or end-to-end latency test, and it does not guarantee stage readiness. Use a controlled soundcheck and follow the warnings from your interface, PA, and venue.

Local karaoke planning (early foundation)

After importing a local track, the current Karaoke session draft can retain your title, manual key, tempo note, and host notes on this PC. It does not fetch, scrape, or bundle lyrics; it does not start playback, change routing, or control displays, remotes, MIDI, queues, or venue screens. Those capabilities remain behind later safety and licensing gates.

Core how-to

The main things you'll do every session, start to finish.

Pick a music style

On the main screen, choose a style preset(pop, rap, rock, R&B, and more). Each one sets the auto-tune feel, tone, and effects to match that genre, so you don't have to touch a single knob to sound right.

Sing and hear auto-tune

Put on your headphones and sing. You'll hear your voice tuned in real time. Use the main auto-tune control to dial it from a natural touch-up all the way to a hard, roboticsnap. There's nothing to press first — what you hear is already corrected.

Adjust in the Mix tab

Open the Mix tab to balance everything: your vocal level, the music, reverb and delay, and how strong the correction is. Small moves go a long way — a touch of reverb and a slightly lower music level usually makes your voice sit perfectly on top.

Voices, personas & the Voice Lab

The Voices panel gives you 57 voices in one tap, organized by vibe: 34 character voices (monsters, fantasy, sci-fi, funny) and 23 broadcast personas(podcast, trailer, sports announcer, villain, and more). Use the search bar to jump straight to a voice — try "dragon" or "radio".

🎨 Voice Lab — make your own. Hit CREATE MY VOICE on the Voices page: you stay live while you drag big simple sliders (pitch, grit, echo, even a choir layer), so you hear your new voice take shape in real time. Name it, pick an icon, save — it becomes a one-tap card under MY VOICES, right next to the built-ins. Start from any preset you love, or from scratch. Free includes 2 custom voice slots; PRO supports up to 60.

The Studio: Beat Maker, pads & keys

Open the Studio to make music behind your voice:

  • Beat Maker — perform on 12 drum pads or build with four velocity levels, mute/solo tracks, four pattern banks, swing, 14 starter grooves, and four synthesized kits. Undo, rotate, then export 1–8 bars as WAV or velocity-aware MIDI.
  • Pads — trigger sounds and loops live while you sing.
  • Keys — a playable piano with 8 instruments and smart auto-chords, so it sounds musical even if you've never played.

Show Mode: setlists & teleprompter

Show Mode is built for performing. Load a setlist, then use the giant NEXT / PREV buttons to move between songs. Lyrics scroll in a teleprompter so you never lose your place, and a TALK button switches to a natural speaking voice between songs — the crowd never hears the seam. It runs fine with no power grid or Wi-Fi.

Record a take & export an MP3

Hit Record and sing. Every take saves three things: the full mix, your polished vocal, and the raw mic. When you like a take, choose Export → MP3 (or WAV) to save a shareable file to your PC.

Make a share clip & score card

After a take, trim a clipof the best moment and share it anywhere in seconds. I Can't Sing also builds a score card from live scoring — your on-pitch %, streaks, and grade — so you can post the clip with your score attached.

Save & load a full song

Working on something bigger? Save the whole project — your vocal takes, beat, keys, and mix settings — and load it later to pick up exactly where you left off. Nothing is locked to the cloud; the file lives on your PC.

Warm-ups

The Warm-ups section runs quick guided vocal exercises before you sing or perform. A couple of minutes loosens up your voice, helps you hit notes more accurately, and pushes your live score higher.

Bring your song from anywhere

The takeover engine works on any audio — an AI song, a DAW bounce, or any file on your PC. Export the track, load it into the Song Deck, and sing the lead yourself. Here's how to get the file out of the common tools.

Suno

Open your song, click the (or Share) menu, and choose Download → Audio (MP3/WAV). Save it to your PC.

Udio

Open the track, use the Download / export option, and pick the audio file (MP3 or WAV).

Riffusion

From the track menu choose Download and save the audio file to your PC.

A DAW

In Ableton, FL Studio, Logic, Reaper, etc., Export / Bounce your mix to a WAV or MP3 file.

Any audio file

Already have a song, beat, or acapella as MP3/WAV/FLAC? That works too — no special export needed.

Load & take over

In I Can't Sing open the Song Deck, drag your file in, let it split the stems, mute the original lead, and sing.

Once the song's loaded, the stem engine separates lead vocal, backing vocals, and instrumentalright on your machine. Mute the AI lead, keep the harmonies (or go full acapella), and the auto-tune locks to the song's real melody while you perform it. Suno is just one example — every source is treated the same.

Go live / streaming

To send your tuned voice into OBS, Discord, Zoom, or any app, you route it through a free virtual audio cable. Set it up once and every app can hear your I Can't Sing output.

1. Install VB-CABLE

Download VB-CABLE (a free virtual audio cable) from vb-audio.com, unzip it, right-click the installer, choose Run as administrator, click Install Driver, and restart your PC. This adds a CABLE Input and CABLE Output device to Windows — think of it as an invisible cable between apps.

2. Send I Can't Sing into the cable

In I Can't Sing open Settings → Audio Devices and set the output to CABLE Input. Your tuned voice now flows into the virtual cable instead of your speakers.

To still hear yourself while streaming, turn on “Listen to this device” for CABLE in Windows Sound settings, or use an audio interface with direct monitoring. Otherwise your voice goes only to the stream, not your own headphones.

3. Pick the cable in your streaming app

  • OBS — add an Audio Input Capture source and choose CABLE Output.
  • DiscordSettings → Voice & Video → Input Device → CABLE Output.
  • ZoomSettings → Audio → Microphone → CABLE Output.

That's it — anywhere you can pick a microphone, pick CABLE Output and it hears your live, tuned vocal.

Troubleshooting

The quick fixes for the things people ask about most. Tap a question to open it.

No sound, or my mic isn't detected

Nine times out of ten this is the wrong device selected. Open Settings → Audio Devices inside I Can't Sing and set your microphone as the input and your headphones or speakers as the output — Windows often defaults to the wrong one.

Still nothing? Check Windows Settings → Privacy & security → Microphone and make sure microphone access is turned on for desktop apps. Unplug and replug a USB mic, then reopen the app so it re-scans your devices.

Audio glitches, crackle, or dropouts

Crackling almost always means the buffer size is too small for your PC. In Settings → Audio, raise the buffer size a step at a time (for example 128 → 256 → 512 samples) until the crackle stops. A larger buffer adds a tiny bit of latency but makes playback rock-solid.

Also close heavy background apps (browsers with many tabs, games, video calls) — they compete for the same audio timing.

High latency / my voice feels delayed

Latency is how long it takes your voice to travel through the app and back to your ears. To get it as low as possible:

  • In Settings → Audio, choose the WASAPI (or ASIO) driver instead of the default shared driver.
  • Use a USB audio interface (like a Focusrite Scarlett) with wired headphones — this gets you to roughly 10–20 ms, which feels instant.
  • Lower the buffer size as far as it'll go without crackling (see above).
  • Avoid Bluetooth headphones — they add a lot of unavoidable delay.
The app won't start

First, restart your PC and try again — the first launch after an update can take up to a minute while it sets up voice models.

If it still won't open, we keep a plain-text log that tells us exactly what happened. Open File Explorer and paste this into the address bar:

%LOCALAPPDATA%\VocalForge\startup.log

Email that file to support@icantsing.com and we'll tell you what to fix. We're a small team and we actually read every message.

I forgot my password

On the sign-in screen choose Forgot password and enter the email you signed up with — you'll get a reset link. Your license key is tied to your account, so resetting your password never affects what you've paid for.

If the reset email doesn't arrive, check spam, then email support@icantsing.com from the same address and we'll sort it out.

License & activation

After you buy, we email your license key. Open I Can't Sing, go to Account, and paste it in. It's verified on your device — no constant login required. Once activated, it keeps working offline.

Moving to a new computer, or your key won't take? Email support@icantsing.com and we'll move it for you.

How does offline mode work?

Everything that touches your voice — auto-tune, melody correction, stem separation, the voices, the Beat Maker — runs entirely on your own PC. After the one-time activation, you don't need an internet connection to use the app, and your audio never gets uploaded or streamed anywhere.

The only things that use the internet are optional: signing in on a new device, downloading updates, and buying or upgrading a plan. You can sing, record, and perform on a plane or at a gig with no Wi-Fi.

Still stuck? We're a small team and we actually answer — email support@icantsing.com and we'll help you get singing.