# Gradium — How do you describe a voice? What we learned from designing 16,000 of them

- Company: Gradium (gradium.ai)
- Announced: 2026-11-18
- Category: not stated
- Coverage: not counted
- Announcement: no
- Group: routine
- Source: https://gradium.ai/blog/how-to-describe-a-voice
- Record: https://forck.live/items/37896-how-do-you-describe-a-voice-what-we-learned-from-designing-16-000-of-them
- Subject: Gradium TTS / Phonon

Gradium's voice team is sharing the prompt patterns it uses to describe voices, drawn from designing more than 16,000 voices and keeping about 1,400 of them for its catalog. The company says the use case is often the most informative part of a description, because "for a professional support line" or "narrating a children's bedtime story" implies a set of choices about pace and energy that adjectives alone do not. Pitch is the easiest dimension to get right, while texture words such as airy or breathy work better as a nudge than an exact instruction, since even the voice team cannot reliably tell them apart. Accents land best when named precisely, and for accents the team cannot judge by ear, such as Peruvian or Bolivian Spanish, native speakers from the region confirm the result before a voice enters the catalog. A role alone is not enough: asked for just "A tour guide.", three drafts ranged from about 110 Hz to 260 Hz and did not agree on gender. Gradium says Voice Design creates each voice newly and uniquely, and the model never outputs the voice of a real person.

## Evidence

Verbatim from https://gradium.ai/blog/how-to-describe-a-voice:

> To build the Gradium voice catalog, we designed more than 16,000 voices from written descriptions and kept about 1,400 of them . Each of those voices started as a few lines of text. Much of what we know about describing a voice comes from writing those lines and listening to what came back.
> 
> Describing a voice turns out to be harder than it sounds. When you design a voice and start describing it in a prompt, the accent and the gender usually come to mind first. Other qualities shape a voice just as much, but they're easy to forget or hard to put into words. Age is one of them. Pitch, texture and tone are others, and it isn't obvious whether words like "calm" or "lovely" mean anything to a model at all.
> 
> It's also a question we hear often from teams building with Voice Design: how do you write a prompt that gets the voice you have in mind? Many of them share their prompts and feedback with us, and each one helps us understand how people describe voices and keep improving the model.
> 
> The gap between the voice you hear in your head and the one that comes back usually comes down to a few words.

---

Record: https://forck.live/items/37896-how-do-you-describe-a-voice-what-we-learned-from-designing-16-000-of-them
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md
