Follow your own pace
Your spoken content updates your place. There is no preset scrolling speed to keep up with.
Meet TalkFollow. Let your script follow your delivery.
A real-time teleprompter that runs on your computer. It finds your place from what you say and shows what comes next, so you can focus on your delivery.
You speak. Your script follows.
Instead of scrolling at a preset speed, TalkFollow matches local speech recognition results to your script and follows your delivery.
Interface preview · View only. Download and install to try the features.
Introducing TalkFollow
Floating view
Live recognition—
Floating viewTalkFollow
- TalkFollow is a real-time teleprompter that runs entirely on your computer.
- It connects your prepared script with your spoken words, helping you stay focused and speak with confidence.
- Import a local Markdown script to keep its headings, text, and formatting in the reading view.
- Choose a local input device and start your session with a microphone.
- Local speech recognition listens continuously and helps the script follow your reading position.
- Pauses, repetition, and off-script remarks are handled carefully to avoid moving ahead too soon.
- You can choose a new reading position manually and continue at your own pace.
- Default, Fixed, and Focus display modes suit different reading preferences and situations.
- Adjust text size, line spacing, and follow colors to make longer sessions comfortable.
- A separate floating view keeps your script alongside your work and can stay on top.
- Optional terminology provides context for names, products, and technical terms in your session.
- Save sessions and organize them into groups so your next presentation is easier to prepare.
- Speech recognition and script matching run on your computer; scripts and models are not uploaded automatically.
- From a short practice to a formal presentation, TalkFollow helps you express yourself naturally.
See your place and what comes next
Different colors distinguish completed, current, next, and unread content.
Take over when you need to
In the app, pause your session or reposition using the current display mode, then continue speaking.
During silence, off-script explanations, insufficient recognition evidence, or repeated or similar content whose position cannot be safely distinguished, the app keeps the confirmed position to avoid incorrect advances. Continue with matching script content, or reposition manually using the controls supported by your display mode. Repeated content does not always prevent Sequential following; the position must still meet the confirmation conditions.
Choose how to follow and how to read.
Follow modes control where the app looks for a match. Display modes control how your script appears. They serve different purposes.
Three follow modes
Choose a matching range based on whether you read in order, skip ahead, or return to earlier content.
CurrentNext unit
Sequential
Default choiceMatches the current unit and the immediately following unit to advance in order. A newly opened session starts at the first matchable unit unless repositioned manually.
When to use it
Record a scripted video or explain a lesson step by step.
CurrentLater script
Forward
Matches from the current unit to the end of the script, allowing you to skip later content. Beta mode.
When to use it
Skip parts of your script as the situation changes, then continue with later content.
EarlierCurrentLater
Free
Matches content before or after your current position, letting you revisit earlier material or jump ahead. Beta mode.
When to use it
Revisit earlier material, respond to audience questions, and adapt your speaking order.
When script analysis finds exact duplicate matchable body units, or a whole unit contained in another at positions allowed by complete-word and terminology boundaries, only Sequential is supported, not Forward or Free. Follow the analysis message. Preparing a Forward or Free session also requires a quick follow speed check on the current device. If it fails, switch to Sequential as prompted and try again. The app does not change your follow mode automatically.
Three display modes
Keep the full script in view, or focus on the content immediately ahead.
Default
Keeps the full script. When the next unit reaches the page end or cannot fit, the view advances with the current unit at the top.
Fixed
Keeps the full script, placing current and next content near the middle of the reader to reduce eye movement.
Focus
After starting, shows only the current and next matchable units. Use the wheel while running; switching units is unavailable while paused.
Display modes do not change follow modes. Click matchable body text in Default or Fixed. In Focus, click before starting and use the wheel after starting. Repositioning is unavailable while paused in every mode. Adjust font size, line height, and highlight colors in Display & language.
Start with your script. Keep it ready for next time.
Import a local script and open a temporary session or save it for later. Check the content as needed; your original file stays unchanged.
Bring your own script
Import common script formats; the app reads them automatically. We recommend checking the preview. Choose a follow mode and optional terminology in the session form; built-in models are automatic. Empty scripts or scripts without matchable Chinese or English body text show a message and cannot be saved.
- Markdown
- TXT
- DOCX
Manage sessions your way
Open a temporary session for a quick practice, or save it to reopen from your session list and organize it into groups.
- Temporary sessions stay out of the saved list
- Save sessions to use them again
- Organize by group and search by title
PDF import supports unencrypted, text-based documents only. Scanned or image-only PDFs are not supported. Text is extracted without the original layout; check the preview.
Make different kinds of speaking feel easier.
Keep your script where you need it, help address recognition errors involving specialist terms, and use the fixed built-in speech models delivered with the app.
Floating teleprompter
Show your script in a separate window with always-on-top and background opacity settings.
- A separate window alongside recording or presentations
- Configurable always-on-top behavior and background opacity
Terminology entries
Add names, brands, and specialist terms from your script to help address recognition errors and better match speech with your script.
- Organize preferred spellings and alternative forms of the same term
- Manage categories and choose enabled entries when creating a session
Local speech models
Version 0.1.0 uses the built-in primary and fixed auxiliary speech models delivered with the app. No separate model installation is needed.
- Chinese and English speech following
- Built-in models work automatically with no separate installation
Your voice and scripts stay on your computer.
Speech recognition, script matching, and following run on your device. Audio, scripts, and terminology are not uploaded automatically. Built-in models ship with the app and need no download during use.
- No voice or script uploads
- Built-in models work offline
- No account required