What is Multi-Speaker Text to Speech?
Multi-speaker text to speech voices a whole conversation: each speaker in the script gets their own AI voice, in one audio file.
The Simple Definition
Multi-speaker text to speech (multi-speaker TTS) converts a script with two or more speakers into a single audio file, giving each speaker a distinct voice. Instead of generating each line separately and stitching them together, the conversation is performed as a conversation, with natural turn-taking and reactions.
How the Script Is Formatted
Each line starts with the speaker's name and a colon. Every distinct name gets its own voice. Plain Latin letters for names work best (Lucia rather than Lucía).
Why It Sounds Better Than Separate Lines
What It Is Used For
A two-speaker script
Host: Welcome back! Today we're asking one question: is coffee bad for you? Guest: [laughing] I was hoping you wouldn't ask me that before my second cup. Host: <short pause> So... is it? Guest: For most people, three or four cups a day is fine. The sugar is the bigger problem.
Two names, two voices. ZenMic casts a voice for each speaker, or you pick them yourself.
Frequently Asked Questions
How many speakers can one audio have?
ZenMic's AI writer casts 1 to 6 speakers, and if you write your own script every named speaker gets a voice.
Can I choose the voice for each speaker?
Yes. Pick any voice for each speaker, including your own cloned voice.
Can speakers use different languages or accents?
Yes. Each speaker can have a voice with a different accent, for example a British host and an Irish guest.
Is it free to try?
Yes. Sign in and make your first audio free, no credit card needed.