Voice-over for video
A script becomes an audio track you download and drop into your own editing software.
CAPABILITY
Text to speech online with AI voices
You paste in the text, pick a voice, and get back an audio file you can play and download. The catalog holds 187 voices grouped by language, English included, and you choose the file format before the run. The tool sits in its own Sound section.
- 187
- voices in the catalog
- up to 50,000
- characters per run
- 4
- file formats
How it works
-
Paste your text
The field takes your whole prepared text in one go. The character limit is set by the model you pick: 50,000 on some, 4,096 on others, and the length is checked before the request is sent.
-
Pick a voice
The catalog holds 187 voices across 14 language groups, plus a separate set of 27 voices with a choice of 20 languages. Language and Turbo mode are set separately on models that support them.
-
Set the format and run it
MP3, WAV, FLAC or OGG, with MP3 as the default. A cost estimate is shown before the run; if your balance is too low, the request is not sent and the service shows how much is missing.
-
Collect the file
Play the finished recording in the built-in player and download it. The run stays in the section history, where you can replay it, repeat it with the same settings or delete it.
What it does
- Input
- The text you paste into the field
- Text limit
- Up to 50,000 characters on some models and 4,096 on others; where a model sets no limit of its own, a 5,000-character fallback applies
- Voices
- 187 in the catalog, plus a separate set of 27 voices
- Languages
- 14 groups: English, Russian, Spanish, Portuguese, French, German, Italian, Dutch, Vietnamese, Arabic, Chinese, Cantonese, Japanese, Korean; the separate voice set offers a choice of 20 languages
- Extras
- Explicit language selection and Turbo mode — faster and cheaper — on models that support them
- File formats
- MP3 by default, plus WAV, FLAC and OGG
- Output
- One audio file per run, with a built-in player and a download button
- History
- Every run is kept in the section: replay, repeat with the same settings, delete
- Where to find it
- The Sound section in the web app, Sound mode in Studio, the Telegram mini app
What people use it for
Podcasts and audio versions of articles
An article or a transcript is voiced in one pass: some models take up to 50,000 characters, so long material goes through in a single run.
Course and training material
Each lesson is a separate run. For the next part you can pick the same voice from the catalog, or repeat the run with the same settings straight from the history.
Voices in other languages
The catalog is sorted into 14 language groups, and the separate voice set offers a choice of 20 languages. You prepare the text in the language you need yourself.
Frequently asked questions
Which voices and languages are available?
The catalog holds 187 voices across 14 language groups, with English, Russian, Spanish, German, French, Chinese and Japanese among them. A separate set of 27 voices adds a choice of 20 languages. You pick the voice before the run.
How much text can I voice at once?
The limit is set by the model you pick: 50,000 characters on some, 4,096 on others; where a model sets no limit of its own, a 5,000-character fallback applies. The length is checked before the request is sent.
What does text to speech cost?
It depends on the model and the amount of text. An estimate is shown in the interface before the run, and the exact amount is fixed once the file is ready. Current rates are on the pricing page. If your balance is too low, the service shows the amount that is missing and the request is not sent.
What format does the file come in?
MP3 by default; before the run you can switch to WAV, FLAC or OGG. The finished recording plays in the browser and downloads with one button.
When am I charged for a voice-over?
Only after the file is ready. The wait is capped at ten minutes: if nothing arrives in that time, the run is marked as failed and the money stays on your balance.
What can I do with the finished recording?
Play it in the built-in player, download it with one button, and mix it in your own editing software. The run stays in the Sound section history, where you can replay it, repeat it with the same settings or delete it.
Related capabilities
Try it on your own task
A starting balance is already on your account, no card required. You pay only for what you generate.