Write your text, pick a voice and get a ready audio track for a video, an ad, a podcast or a course. Go to text to speech →
Speech synthesis, your own voices and transcription in one place.
Paste your text, pick a voice and get a ready mp3. Long text is voiced in parts and joined into a single track.
English, Russian and dozens of other languages. The language is detected from the text itself.
Read a few phrases out loud, then voice any text with your own voice.
Describe it in words, for example a warm male narrator, and the service creates that voice and saves it to your list.
You can voice a text right inside the talking avatar form: a photo plus a voice make a talking video.
Instead of typing you can dictate: the service transcribes your speech into the text field.
Type it, paste it or dictate it.
From the catalog, your own, or one created from a description.
From 0.75× to 2×. You see the price before you start.
Listen in the browser, download the mp3 or use it in a video.
Two ways to get a voice that is not in the catalog.
The service shows a few phrases, you read them out loud, and the voice is built from that recording. You can listen to the recording before sending and record it again.
Describe the gender, age, tone and manner of speaking, and the service offers two options to choose from. No real person stands behind such a voice.
You may only clone your own voice. Before creating it you confirm that you are recording yourself.
The voice reads what is written. Two tools help it read the text the way you meant it.
Below the text field there are buttons: pause, whisper, emphasis, emotions, laughter. A button inserts a short note in square brackets — the voice does not read it aloud, it changes the intonation of the next phrase or makes a pause.
AI writes numbers, dates, amounts, abbreviations and web addresses out as words, and foreign words in the letters of the text’s language, so the voice reads them correctly. In the second mode you pick a speaking style — calm, energetic, businesslike or your own — and AI adds intonation and pauses for you. You can compare the original and the new text and choose which one to voice. The price is shown on the button before you start.
A talking avatar needs a ready audio track. You do not have to record it yourself: the avatar form has a voice-over button. Write the line, pick a voice, and the track is filled in together with its duration. More about the talking avatar
| Action | Credits |
|---|---|
| Speech, 1000 Latin characters | ··· cr. |
| Speech, 1000 Cyrillic characters | ··· cr. |
| Clone of your own voice | ··· cr. |
| Voice from a description (two options) | ··· cr. |
| Dictation, per minute of audio | ··· cr. |
The price depends on the text length: a Cyrillic letter takes twice as much space as a Latin one. The exact price is shown in the form before you start.
Yes. In the Voice section start creating a voice and read out loud the phrases the service shows. Your voice is built from that recording, and after that you can voice any text with it.
No. A clone can only be made from your own voice: before creating it you confirm that you are recording yourself. This is both a legal requirement and protection against voice fraud.
The service splits a long text into parts and joins them into a single track, so a large text takes longer.
The credits return to your balance automatically. The reason is shown on the voice-over card.
In the Voice section: you can replay them, download the mp3, repeat with the same text or delete them.
Paste your text and pick a voice. The finished track usually appears in under a minute.
Go to text to speechWe use cookies to operate the service, keep your session, and collect anonymous statistics. See our Privacy Policy.