Voice Factory TTS Create audio
Voice Factory TTS › Use cases › Dialogue audio

Dialogue audio with Script mode

Sometimes you need people talking to each other rather than one voice narrating — an audio drama, a conversation textbook, a workplace-safety role-play. Script mode reads each line in a different voice and joins it all into one file, so there is no stitching recorded fragments together in an editor.

1. How to write the script

At the top right of the text box, press Script (several), then write one line of dialogue per line as Name: line.

Basic formNarrator: Monday morning, and two people have arrived at the office early. Minsu: Jieun, did you have a good weekend? Jieun: I did! I went hiking for the first time in ages. Minsu: Lucky you. I just slept at home.

2. Giving each character a voice

As soon as you type a script, a Character voices list appears below. Female and male voices are assigned alternately to start with, and the Narrator begins on a calm male voice. You can change voice, tone and effect per character, and ▶ plays that character's first line.

When you have more characters than voices, tell them apart with tone and effects.

CharacterSuggested setting
A childFemale voice + "Cute" tone, or the "Child" effect
An elderly person"Calm" tone + "Elder" effect
A robot or AI assistant"Robot" effect
A voice on the phone"Phone" effect
A giant or monsterMale voice + "Giant" + "Echo"
A public announcement"ARS" or "News" tone + "Echo"

3. The gap between lines

Gap between lines under Script settings is the silence inserted between one line and the next. Everyday conversation sounds natural at 0.3–0.5 seconds; for teaching material where the learner repeats after the speaker, raise it to 1.5–2 seconds. If you only want breathing room at scene changes, split the narration with commas instead.

4. Worked examples

An English conversation lesson

Assign "multilingual" voices (Ava, Andrew and the like) to the characters and the English lines are read naturally. Leave the Narrator on a Korean voice and one file can hold both the Korean explanation and the English dialogue. A gap of about 1.5 seconds gives the learner time to repeat.

Example scriptNarrator: Ordering coffee at a cafe. Staff: Hi, what can I get for you today? Customer: Can I have an iced latte, please? Staff: Sure. Would you like it small or large? Customer: Large, please. Narrator: They ordered a large iced latte.

A workplace-safety role-play

Showing an incident as a conversation sticks far longer than reading out the rules. It works well as audio dropped into training slides.

Example scriptNarrator: During a server-room inspection, a new employee is about to work on the cabling. New hire: Should I just pull this cable out? Senior: Hold on — what do we check before any work starts? New hire: Ah… isolate the power, and put the warning sign up. Senior: Right. The moment you skip the check is the dangerous one.

Bedtime stories and audio drama

Put the Narrator on the "Gentle" tone and lay "Calm music" underneath at about 15% and it takes on a bedtime-story feel. Giving each character a different effect lets children tell who is speaking from the voice alone.

5. Subtitles come too

Files made in Script mode also have a CC button for subtitles (SRT). The gaps between lines are accounted for, so dropping them onto a video keeps dialogue and captions aligned. Only the spoken lines go into the subtitles; the character names are left out.

When it does not work

In Script mode, press "Insert sample" for a short example conversation. Listen first, then change the names and lines.

Create from script

Related guides

Making an audiobook with TTS: from manuscript to chaptersHow to turn your own writing, or a public-domain work, into a TTS audiobook. Lecture and talk audio: slide narration and in-house trainingHow to make narration for online lectures, staff training and presentations. Store announcements: closing time, flash sales, parking noticesHow to make announcements for shops, cafes, stores and events with TTS.