Download Speechius
Speechius is a desktop prompting tool that listens to spoken words and moves the script accordingly. It is designed for recording tutorials, hosting webinars, delivering presentations and speaking on video calls. The window can remain above other applications, and capture-exclusion mode keeps it out of common screen-sharing and recording paths. Projects group multiple scripts, autosave preserves edits, and .speechius files transfer complete setups between installations. The application runs on macOS 13 or later and Windows 10/11, with recognition and interface in English and German. A 14-day trial is available, and paid licenses are one-time purchases with 12 months of updates.
The software uses automatic speech recognition to compare live speech with the upcoming script text. When a match is found, the reading position advances, so the speaker does not chase a fixed scroll rate. Clicking any word re-anchors the cursor after a detour. Slide triggers advance presentation slides when trigger words are spoken, keeping hands free and eyes on the camera. Adjustable transparency, text size and reading line adapt the prompter to different screens and camera positions. A live transcript shows the last recognized words, so tracking can be verified at any moment.
Speechius is distributed for macOS and Windows as a one-time purchase without a subscription. The Pro edition covers one person on up to two devices, and the Studio edition covers one company with up to five seats. Both editions include the full feature set and 12 months of updates. After that period, the purchased version continues to run. Speech recognition works locally on Windows with bundled models, while macOS requests Apple's on-device recognition first. The interface and recognition support English and German, and a full 14-day trial is available without a payment card.
Use Cases for Speechius
| Speech-to-script matching advances reading position automatically | Click re-anchoring restores cursor after script deviation |
| Project grouping with automatic persistence of edits | Last reading position restored when switching scripts |
| Plain text import and full project transfer between installations | Backup and migration of triggers and skip markers |
| Window-exclusion API prevents prompter appearing in screen captures | Compatible with conferencing and streaming capture paths |
| Bundled on-device recognition models for offline operation | No audio or script upload to vendor servers |
| Spoken trigger words dispatch slide advancement commands | Hands-free presentation control during delivery |
| Prompter window layered above all other applications | Script remains visible over slide decks and calls |
| Recognition and interface languages selectable independently | Mixed-language configurations for speaker preference |
| Opacity reduction from fully visible to nearly transparent | Desktop and participants visible behind prompter text |
| Word-level cursor placement and Shift-scroll navigation | Manual correction when recognition drifts off script |
| Draggable anchor line aligns gaze with camera lens | Adjustable vertical resting position for reading |
| Text scaling from fifty to two hundred percent | Highlight and anchor remain aligned at every size |
| Toolbar readout displays last recognized words | Immediate feedback on tracking activity and drift |