Dubformer Studio Playbook

Voices

Voices for a project are pulled from the pool of voices labeled Suggested. Selection is automatic: the system rotates through that pool so consecutive videos don't reuse the same voice.

How do I stop a voice from being used in a project?

If you don't like a voice and don't want it selected again:

  1. Go to the Voices tab.

  2. Select the voice you want to stop using.

  3. Click its Suggested label and remove it.

How do I find other voices?

All voices are multilingual β€” the same voice can speak multiple languages. When choosing one, look at its accent and description, not just its source language:

  • Standard accent β€” speaks with little to no noticeable accent, so it works well for dubbing into multiple languages.

  • Regional/named accent (e.g. Parisian) β€” carries a distinct accent. That accent will typically carry over even when the voice is used to dub into a different language.

To add a voice to your selection pool, add the Suggested label to it.

We recommend previewing a voice using text similar to your actual content rather than the default sample text β€” use Customize sample text for this.

If your language has enough S-tier voices available, prefer those. S-tier is our highest-quality voice model.

Before using a voice, confirm it's marked Without live moderation. This matters especially for news content: voices with live moderation have a list of Prohibited topics they can't be used for.

Text & Script Formatting

You can control how a voice delivers a line β€” emotion, pacing, emphasis, pronunciation β€” directly from the script text.

How do I control emotion, delivery, or pacing?

Use audio tags: bracketed instructions inserted directly into the text.

  • Emotion/voice: [laughs], [laughs harder], [starts laughing], [wheezing], [whispers], [sighs], [exhales], [sarcastic], [curious], [excited], [crying], [snorts], [mischievously]

  • Direction/reactions: [shouts], [clears throat]

  • Sound effects: [gunshot], [applause], [clapping], [explosion], [swallows], [gulps]

Example: [whispers] I never knew it could be this way...

How do I control pauses and rhythm?

Use punctuation or audiotag

  • A hyphen - or em dash β€” creates a pause. Chain them (-- --) for a longer pause.

  • An ellipsis ... adds a pause plus a hesitant, nervous tone.

  • An audiotag [short pause], [long pause]

How do I stress or emphasize a specific word or syllable?

There's no dedicated "stress" tag. Use instead:

  • Capital letters β€” the main way to mark stress

  • Use stress markers to place emphasis on a syllable: primary stress ˈ, secondary stress ˌ.

    Apostrophes or single quotes around the stressed letters

Example: to stress the "ii" in trapezii, write it as trapezIi.

How do I get precise pronunciation for a tricky word?

Use IPA (International Phonetic Alphabet) β€” "/ˌbaΙͺoʊˈkemΙͺstri/"

What if the voice mispronounces a word or name?

If speech synthesis mispronounces an unfamiliar word β€” a foreign name, for example β€” you can respell it phonetically, "as it sounds."

How do I emphasize a different word in a sentence?

This is about logical stress β€” which word in the sentence carries the meaning, e.g. "It WAS my sister" vs. "It was MY sister."

If the logical stress lands on the wrong word, try emphasizing the intended word by capitalizing it or putting it in quotation marks.