Observed arrival · 2026-10-26
Freya asks your ear to judge its synthetic voices
Freya pairs an English text-to-speech demo with a blind listening test that asks visitors to identify the human voice.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- Developers and researchers evaluating speech synthesis
- Worth noticing
- Its displayed Elo table puts human speech at 1462 and Freya Adam V1 at 1419, alongside selected competing models.
Field notes
The blind test has visitors hear two clips of the same sentence and choose which sounds human; the page says its ranking uses Elo and labels the table as selected models, measured 2026-09-16. The live generator includes speed controls and a 300-character input counter. Separate research links lead to a Turkish text-to-speech paper, model, code, and benchmark, with the model labeled Apache 2.0.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
⊠LoginAccess appeared gated
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
◎NicheUnusually specific use
ƒJavaScriptBrowser-side code central
One card from the complete issue