Voice Training for Personalized Scripts
Overview
Voice training helps Teleprompter learn how you naturally speak so generated scripts can better match your tone, word choice, and reading level. Training is optional; you can complete it during first-time onboarding or later in Settings.
Before You Start
- Microphone permission is required for speech-to-text while you record a sample.
- You can store up to 20 voice samples. Three or more samples are enough for strong personalization; additional samples have diminishing returns.
How voice training works
The app does not store your raw voice recording. It uses speech-to-text to capture what you said; that text is saved to personalize script generation. The in-app disclaimer explains this before you record.
How to train your voice
- Open Settings and find the Voice Training section (or complete the Voice Training step during onboarding).
- Expand Train Your Voice if the section is collapsed.
- Choose a suggested topic or enter a custom topic.
- Tap the microphone control and speak naturally — as if explaining the topic to a patient, family member, or caregiver.
- When you finish, confirm the sample. You can add more samples or delete an existing sample from the list.
Troubleshooting
| Issue | Cause | Fix |
|---|---|---|
| “Limit Reached” for voice samples | 20 samples stored | Delete a sample you no longer need, then add a new one. |
| Microphone or speech recognition errors | OS permissions or noisy environment | Enable the microphone for Teleprompter in device Settings; try again in a quieter space. |
FAQ
Q: Is voice training required?
A: No. Onboarding explains that you can skip and train later in Settings.