01 / WRITE FOR THE EAR
Prepare text that sounds natural when spoken
A voiceover script is not the same as a page of written copy. Listeners cannot scan backward, so use direct sentences and introduce one idea at a time. Read the script aloud before generating it. If you run out of breath or lose the point, shorten the sentence.
Punctuation controls more than grammar in text-to-speech. Periods create clear stops, commas suggest short pauses, and paragraph breaks separate ideas. Spell out uncommon abbreviations, write numbers the way they should be spoken, and check product names or technical terms with a short preview.
- Remove URLs, notes, and formatting that should not be spoken.
- Expand uncommon abbreviations.
- Use punctuation to guide pacing.
- Break long scripts into logical sections.
- Confirm that you have permission to use the source text.
02 / GENERATE
Create an MP3 in TTSHub
- Add the text.
Paste a script into TTSHub, or upload UTF-8 TXT files when you need several voiceovers from a prepared batch.
- Choose language and voice.
Filter the voice list by language or region. Select a voice that fits the audience and tone rather than choosing only by novelty.
- Set quality and speed.
Use the available speech model and pacing controls. Begin near the default speed; extreme values can reduce clarity.
- Preview a difficult sentence.
Test names, abbreviations, dates, prices, and technical phrases before generating the complete script.
- Generate and download.
Follow the progress, listen to the completed audio, and save the MP3. Batch jobs can be collected as individual files or a packaged download when available.
03 / REVIEW
Listen for meaning, not only pronunciation
Review the MP3 from start to finish. A word can be pronounced correctly while the sentence still feels rushed, flat, or ambiguous. Listen for missing pauses, an incorrect emphasis, inconsistent volume, and phrases that sound unnatural when heard aloud.
When a section fails, adjust the text before changing the entire voice. Add punctuation, replace a difficult abbreviation, split one long sentence into two, or rewrite a number phonetically. Generate only the revised section when your production workflow allows it.
| Audio issue | Script adjustment |
|---|---|
| Sentence feels rushed | Add punctuation or split it into two sentences. |
| Name is mispronounced | Try a phonetic spelling in a short preview. |
| Long list sounds confusing | Use shorter items and stronger pauses. |
| Tone feels too fast | Reduce speed slightly and simplify the copy. |
04 / ADD TO VIDEO
Move the finished voiceover into PeteCut
After generation, use the option to add the completed audio to PeteCut when it is available, or download the MP3 and import it into your project. Place the voiceover on the timeline first, then arrange visual clips around its structure. This keeps scene changes aligned with the narration.
Once the edit is stable, PeteCut can generate captions from the timeline audio. Review the resulting caption text and timing before export—the voiceover script helps, but automatic transcription can still interpret names or punctuation differently.
05 / QUESTIONS
Text-to-MP3 questions
Which text format works for batch generation?
Use clean UTF-8 TXT files. Keep one logical script in each file and give every file a descriptive name.
How can I make text-to-speech sound less rushed?
Shorten sentences, improve punctuation, add paragraph breaks, and make a small speed adjustment after testing the script at its default pace.
Can I edit the MP3 in PeteCut?
Yes. Import or hand off the generated file, place it on the timeline, and combine it with video, images, text, and captions.