Text-to-speech with lifelike voices, 30+ languages including Russian.
MiniMax is available in IIshki Studio — audio generation from a prompt, paid for with tokens. IIshki club members get reduced prices on every model. Pick a model above, set the parameters and start the generation.
MiniMax Audio is an AI text-to-speech model with lifelike voices in 30+ languages, including Russian. In IIshki Studio it runs online: paste up to 5,000 characters, pick a voice and get a ready MP3 or WAV for video, ads, tutorials and podcasts.
Open MiniMax in IIshki Studio, paste your text (up to 5,000 characters), pick a voice and press Create. The language is detected automatically, or you can set it yourself. In a few seconds you get an MP3 or WAV file to download; if Telegram is linked, it arrives there too.
Eight Russian voices — four male and four female, from confident to dramatic — plus an English narrator. You can listen to each one above or in the studio before generating.
The editor lets you pick an emotion for the whole text and insert a pause of any length — in the prompt it looks like <#0.5#> (half a second). Speech speed is set separately: slow, normal or fast. Punctuation works too: a full stop pauses longer than a comma, a question mark changes the intonation.
Speech 2.8 HD gives the best quality and the most natural sound — the default choice. Speech 2.8 Turbo is faster and cheaper with a slightly simpler sound, handy for drafts and long texts. Older generations (2.6 and 02) are kept for compatibility.
MiniMax is quick, affordable voice-over with good Russian; ElevenLabs offers more voices, languages and fine emotion control. Try both on the same text and pick the sound you like.
The price depends on the length of the text and is shown in tokens on the model card (1 token = 1 ₽ ≈ $0.011). Tokens can be bought without a subscription; club members get the club price.