01 / PLAN
Start with the result and the source
Generate the voiceover in logical sections when you expect revisions. Smaller files make it easier to replace one sentence without recreating the full narration.
Write for listening rather than silent reading. Short sentences, deliberate punctuation, and a quick pronunciation test usually improve a voiceover more than an extreme speed adjustment.
02 / WORKFLOW
Complete the workflow step by step
- Prepare.
Generate and review the voiceover in TTSHub, then use the PeteCut handoff or import the MP3 manually.
- Set up.
Place narration on the timeline first, trim unwanted silence, and leave intentional breathing room between ideas.
- Create.
Arrange video and images around the spoken structure, then lower background music beneath speech.
- Review.
Generate captions from the stable timeline, correct the transcript, and export only after a complete playback.
03 / QUALITY CHECK
Review the result before the next step
Listen from beginning to end with headphones. Check names, numbers, pauses, volume, and whether the delivery still communicates the intended meaning.
- No word is clipped at an edit point.
- Visual changes support spoken ideas.
- Music never masks important speech.
- Captions follow the final audio timing.
04 / QUESTIONS
Common question
Can I replace one TTS sentence later?
Yes. Generate the revised sentence or section, insert it on the timeline, and recheck spacing, volume, and captions around the edit.
Can I update this workflow later?
Yes. Save the source and project details, then update the relevant settings, content, or output when the destination requirements change.