Text to speech
Some text is easier to hear than to read: a draft you want to proof with fresh ears, an article for the commute, notes you would rather absorb while doing the dishes. Paste up to 4,000 characters below, pick one of three voices, and press play. When it sounds right, download the .mp3. No account needed to try, and nothing you paste is kept.
0 / 4,000 characters
Good uses for a quick read-aloud
Hearing your own writing is the fastest proofread there is: clunky sentences and missing words announce themselves the moment a voice hits them. It also works the other direction, turning reading into listening. Convert a long email before a drive, a study guide before a walk, or the minutes you need to review before a meeting. The download button gives you a plain .mp3, so the audio plays anywhere without fuss.
Frequently asked questions
Is the text to speech tool free?
Yes. Anonymous visitors get 5,000 characters to try, no account needed. A free account (no credit card) raises that to 50,000 characters a month, and paid plans go much higher. When you reach the cap the tool tells you how much you have used and, on monthly plans, when the allowance resets.
What voices can I choose?
Three. Aria is warm and conversational, the everyday default. Atlas is deep and steady, which suits narration and longer reads. Nova is bright and energetic. Pick whichever fits the piece; switching voices and replaying costs nothing extra beyond the characters themselves.
Can I use the audio commercially?
Yes. The audio you generate is yours to use, including in commercial work: videos, podcasts, course material, product demos. Download the .mp3 and take it with you.
Which languages are supported?
The current voices speak English. Wider language coverage is on the roadmap, but for now text in other languages will be read with an English voice, which rarely sounds good.
Why is there a 4,000-character limit?
Synthesis is fastest when the request is short, so each read is capped at 4,000 characters (roughly 600 to 700 words, or four minutes of audio). For longer pieces, split the text at a natural break and stitch the .mp3 files together.
What happens to the text I paste?
It is sent to the speech model to render the audio and is not kept afterwards. There is no reading list of what you have synthesised, and anonymous use is not tied to an identity beyond the free allowance counter.
Or skip the keyboard entirely
tautau puts a mic button next to every text field on the web. Click it, speak, and your words land in the box with punctuation handled. A free account gives you 5 dictations a day, no credit card required.