# AI voice generator

> Text to speech with a curated roster, your cloned voices and Characters that speak: Voice One (3 credits / 100 characters) and Voice Turbo (2). Speed, expressiveness, 13 languages, MP3 out.

Type a line, pick a voice, hear it. The voice studio renders expressive speech on Voice One, fast speech on Voice Turbo, from a curated roster (Aria, Orion), from voices you clone, or from a Character. Punctuation directs the delivery; speed and expressiveness shape it; the result is an MP3.

## Directing a read

- Commas breathe, ellipses hesitate, dashes cut, questions lift — write the performance into the punctuation.
- Expressiveness 0 is steady narration; 1 is theatrical. Speed 0.7–1.3.
- The Delivery guide lists six techniques with example lines you can insert.

## Voices

The roster: Aria, warm and intimate; Orion, low and unhurried, late-night radio. Clone your own voice from a clean minute of audio. Assign any voice to a Character so every line they speak — in the text studio, the API, a film — is the same voice. Thirteen languages on both engines.

## Pricing

Voice One is 3 credits per started block of 100 characters, Voice Turbo 2. A 1,000-character paragraph is 30 or 20 credits; a 55,000-character audiobook chapter about 1,650 or 1,100.

## Starting points

- **Late-night radio** — It's past midnight, and the city's finally quiet. Stay with me a while… we've got nowhere else to be.
- **Ad voice-over** — Thirty seconds. One bottle. Everything you were going to say, before you had to say it.
- **Audiobook opener** — Chapter one. The letter arrived on a Tuesday, which was, she would later say, the first thing that went wrong.

## FAQ

### How natural is the AI voice?

Voice One is an expressive engine: it reads punctuation as performance and takes speed and expressiveness controls. Try a line with the 50 free credits and judge with your own ears.

### Can I clone a voice?

Yes — upload a clean recording in the voice studio. Clones are private to your account.

### What languages are supported?

Thirteen, on Voice One and Voice Turbo.

### What format is the output?

MP3, returned directly — in the studio, from POST /v1/audio/speech, and as a file from the CLI or the Adobe panel.

## Related

- https://eroq.ai/models/eroq-voice-one — Voice One
- https://eroq.ai/tools/ai-voice-cloning — Voice cloning
- https://eroq.ai/use-cases/audiobooks-podcasts — Audiobooks & podcasts
- https://eroq.ai/use-cases/faceless-youtube — Faceless YouTube

---

This page as HTML: https://eroq.ai/tools/ai-voice-generator · Studio: https://eroq.ai/studio · Docs: https://eroq.ai/docs · Machine index: https://eroq.ai/llms.txt