Documentation
How LectureVoice works
Every project moves through the same six-stage pipeline: upload, voice, scripts, audio, render, done. Here's what happens at each stage and how to get the best results.
1. Upload slides
Drop in a PDF or PowerPoint deck (up to 50 MB).
- PDF and PPTX are both supported: PPTX is converted to slide images automatically.
- Once ingestion finishes, pick exactly which slides make it into the video with All / None / First N / Middle N / Last N, or click individual thumbnails.
- Only selected slides move on to scripting, audio, and rendering.
2. Choose your voice
Reuse a saved voice profile or record a new one.
- Recording 20-30 seconds of clear speech gives the cloned voice the best reference quality.
- Voice profiles are saved to your account and can be reused across every future project.
- Preview any saved profile's reference clip before selecting it.
3. Generate scripts
AI drafts a narration script for each selected slide.
- Scripts are generated per-slide from the slide content, so you can review and edit any of them before continuing.
- Regenerate a single slide's script if the wording needs another pass.
4. Synthesize audio
Each script is narrated in your cloned voice.
- Audio is synthesised slide-by-slide; you can preview and re-run any clip that doesn't sound right.
- This step requires every included slide to already have a script.
5. Render the video
Slides and narration are assembled into one MP4.
- Rendering requires every included slide to have both a script and a synthesised audio clip.
- Track progress live: assembly runs as a background job.
6. Done
Download or share your finished lecture video.
- The finished MP4 is available for playback and download from the project's Done step at any time.
- Come back to any earlier step later; projects are saved and resumable from the dashboard.
Tip: record your voice reference somewhere quiet, and aim for natural, continuous speech rather than a word list; it gives the clone more to work with.
Frequently asked questions
What file types can I upload?
PDF and PowerPoint (.pptx) slide decks, up to 50 MB per upload.
Can I only narrate some of my slides?
Yes. After upload, use the selection tools (All / None / First N / Middle N / Last N, or click individual slides) to choose exactly which ones get scripts, audio, and a spot in the final video.
Do I need to re-record my voice for every project?
No: voice profiles are saved to your account and can be reused across any number of projects. Record once, reuse everywhere.
Can I edit an AI-generated script?
Yes, scripts can be reviewed and regenerated per slide before you move on to audio synthesis.
Can I resume a project later?
Yes: every project remembers the furthest step you've reached. Click "Continue" on the dashboard to pick up where you left off.
Ready to create your first lecture video?
Head back to the dashboard to upload your first slide deck.