MotionShine01 / Write
Name each line
Start every line with the speaker and a colon, like Anna: or BEN:. The speakers list fills in as you type.
Free tool
Text to speech with several voices.
Write a dialogue, a story or an ad script with a name in front of each line, give every speaker their own AI voice, and get one audio file with the whole conversation. Set the pause between lines, save MP3 or WAV, and take the subtitles as an SRT file with the speakers' names. The voices run in this tab, so your script is never uploaded.
Speakers
Timing
Lines without a name are read by the Narrator. Words in [brackets] are not read.
This runs in your browser. Your script is never uploaded.
How it works
A cast of voices, one file.
02 / Cast
Give each a voice
Every speaker gets a different voice to start with. Change it, and press Hear to listen to their first line.
03 / Save
One audio file
The lines are joined with the pause you set. Save MP3 or WAV, plus an SRT file to use as subtitles.
Questions
Good to know.
How many speakers can I have?
As many as you like. There are eight English voices, so a cast of up to eight can all sound different; more speakers share voices. Voices in other languages can be mixed in, for a line in Spanish or Hindi.
Is my script uploaded?
No. The AI voices are downloaded to your browser and speak there. The first time, the English voice engine is about 92 MB; it is kept for next time.
Can I use the audio in videos and ads?
Yes. Every voice's licence allows commercial use, and the audio is made on your own computer. Some voices ask for credit, listed on the text to speech page. Check the rules of the platform you post on about AI voices.
What is the SRT file for?
It holds each line with its start and end time and the speaker's name, so a video editor or YouTube can show it as subtitles. Edit it in the subtitle editor if you want to change a line.
Can the voices show feeling?
Only through the words and the punctuation: commas and full stops give natural pauses. There are no emotion controls, and directions like [laughs] are skipped rather than acted.