Updated 15 October 2026 · prices and limits reflect the current service settings
Text-to-speech with ElevenLabs voices, your own voice from a description, voice changing on a recording, sound effects and noise cleanup.
1. Where to find it
In Telegram (Mini App)Create → Audio: AI Voice, AI Sound, Voice changer, My voice; Audio cleanup is in the Tools tab
On griffonhub.ruAI audio menu: Voiceover, Sound effects, Voice changer, My voice, Audio cleanup
2. Voiceover: price and models
The price depends on the model: 4 G to 8 G per 100 characters, at least 5 G per run. The price is shown under the field before you start.
| Model | Strength | Text per run | Price |
|---|---|---|---|
| Eleven v4 | default: the most natural speech, emotion tags | up to 10,000 characters | 8 G per 100 characters |
| Eleven v4 Turbo | half the price of v4, also with emotion tags | up to 10,000 characters | 4 G per 100 characters |
| Eleven v3 | expressive, emotion tags | up to 5,000 characters | 6 G per 100 characters |
| Multilingual v2 | steady narration, speed and style settings | up to 10,000 characters | 6 G per 100 characters |
| Flash v2.5 | fast and half the price — for long texts | up to 40,000 characters | 4 G per 100 characters |
Price examples on Eleven v4
300 characters24 G
1000 characters80 G
5000 characters400 G
3. Voices and settings
15 service voices, all speak Russian:
- Search the ElevenLabs library — thousands of voices by gender, age, accent and style.
- Voice settings: stability and similarity; Multilingual and Flash also have speed and style.
- Emotion tags for Eleven v4 and v3 go into the text: [laughs], [whispers], [sighs], [excited], [sarcastic], [curious], [crying], [shouting], [singing].
4. My voice, voice changer and effects
My voicevoiceover price of the sample × 3describe the timbre, age and manner — you get three variants and keep the best for voiceovers
Voice changer1.6 G per second, at least 20 Gyour recording in another voice, keeping intonation and pauses
Sound effects1.6 G per second, at least 15 Ga sound from a description: footsteps, rain, an explosion, a click
Audio cleanup2.5 G per minute of recording, at least 3 Gnoise, hum and echo go, the voice stays; recordings up to 30 min
5. Getting natural speech
- Write as you would speak: short sentences, punctuation, paragraphs.
- Ellipses and dashes add pauses, an exclamation mark adds emotion.
- Stress (for Russian) — capitalise the stressed vowel: “МосквА”, “интерЕсный”.
- If stress marks don't help, write words the way they sound.
- For a long text, ask any AI to mark stresses with capital letters.
6. Plans
- Audio cleanup is available on every plan, including Free.
- Voiceover, my voice, voice changer and effects — from the Plus plan.
7. Payment, errors and support
- The price is on the start button before anything is charged. Griffs are charged when you start, and if a generation fails they are refunded automatically.
- Finished works are in My works on the site and Works in the Mini App. Files are kept for 30–60 days — download what you want to keep.
- Parallel generations depend on the plan: Free — 1, Plus — 2, Pro — 3, Max — 4.
- Adult content, violence and anything illegal is rejected automatically.