VocalScroll
Launch VocalScroll
Guides Library

Mastering Voice-Controlled Scrolling

Dial in speech recognition so the teleprompter responds to your pace—not the other way around.

Last updated September 29, 2025

How Voice Mode Works

When you enable Voice mode in VocalScroll, the app listens for meaningful keywords from your script. Filler words such as “uh,” “a,” or “and” are filtered out automatically. As soon as a keyword sequence is recognized, the current line is highlighted and the scroll position nudges forward just enough to keep the next sentence in the marker zone.

The recognition engine runs locally in your browser through the Web Speech API. On desktop, Chrome and Edge deliver the most consistent results. Safari on macOS also works, but we recommend adding a short warm-up sentence to help the recognizer lock on to your voice.

Set Up for Reliable Recognition

  1. Pick the right microphone. Under Settings → Microphone select a direct USB mic or audio interface. Laptop mics pick up more echo, which can delay matching.
  2. Reduce background noise. Close unnecessary browser tabs, mute notification sounds, and record in a quiet room.
  3. Warm up the model. Read the first two lines of your script before you hit record. This primes the keyword list and lowers the chance of early drift.
  4. Use punctuation strategically. Short sentences with clear commas and pauses make it easier for the model to reset between thoughts.

Troubleshooting Drift

If the highlight falls behind or jumps ahead, try the following:

How to Calibrate Voice Mode

  1. Warm up the recognizer. Switch to Voice mode, then read the first two lines of your script so VocalScroll learns your pacing.
  2. Dial in sensitivity. Confirm the right microphone is selected and lower scroll sensitivity if you see early jumps or lag.
  3. Record a test run. Capture a short rehearsal, review the highlight alignment, and tap Re-Sync or adjust scroll speed until the script stays centered.

Advanced Tips

When you need to dial in recognition, record a short test take and review the highlighted words to confirm the teleprompter keeps pace. Tweak pacing by adjusting scroll speed and by smoothing sentence breaks with brief pauses.

For creators who re-record the same script often, store alternate pronunciations directly inside the script in parentheses. VocalScroll ignores the parenthetical text visually but retains the keywords for matching. Pair these tweaks with the Pre-Recording Checklist to keep your space optimized for every take.

Voice Control FAQ

Which browsers recognize my voice most reliably?

Chrome and Edge on desktop deliver the fastest keyword detection for VocalScroll. Safari on macOS works once the recognizer warms up, while Android browsers perform best with a wired or USB-C microphone connected.

How can I reduce background noise from triggering scrolls?

Use a directional microphone, keep it within 6–8 inches of your mouth, and close other apps playing audio. You can also lower the scroll sensitivity in Settings so VocalScroll waits for two keyword matches before advancing the script.

Can I switch between voice and timed scrolling mid-session?

Yes. Toggle between Voice and Timed in the Control Panel whenever you need. VocalScroll remembers your last scroll speed, so when you return to Voice mode it resumes listening with the same pacing.

Why Creators Trust VocalScroll

Onitsoga Labs builds VocalScroll and keeps every workflow private, tested on desktop and mobile browsers.