Version types explained
Original, Natural, Lecture, and Conversational document versions.
Each document can have multiple versions: alternate renderings of the same source tuned for different listening styles.
New version
Natural
Smooth spoken rewrite
Lecture
Dense, informative
Conversational
Dialogue style
Version types
| Type | Description |
|---|---|
| Original | Your imported file text (or translation if output language differs) |
| Natural | Rephrased for smooth, natural speech |
| Lecture | Dense, information-focused delivery |
| Conversational | Dialogue or podcast-style presentation |
Feature comparison
| Capability | Original | Natural | Lecture | Conversational |
|---|---|---|---|---|
| Original phrasing | Yes | |||
| Available in many languages | Yes | |||
| Natural sounding phrasing | Yes | Yes | ||
| Removes redundancy | Yes | Yes | ||
| Chapter and section breaks | Yes | Yes | Yes | Yes |
| Clean up document artifacts | Yes | Yes | ||
| Paced for information density | Yes | |||
| Optimized for memorization |
When to use each
- Original: fidelity to source material
- Natural: articles and blog posts that sound awkward when read verbatim
- Lecture: textbooks and technical docs
- Conversational: engaging summaries or dual-voice feel
Credits
Generating a new version uses AI tokens plus a 1 credit flat surcharge. Speechdash shows an estimated credits amount before you confirm. Translate (More menu) is billed like MP3 export instead. See Version generation workflow and What uses credits.