A Tiny Speech Judge That Agrees With My Ears

Sat, 07 Nov 2026 15:00:00 +0300

I am building a very small text-to-speech model, whose decoder (the part that makes the sound) has about 1.75 million parameters, and I want it to sound as good as Kokoro-82M and eventually better. Improving it means trying many variants and keeping the ones that sound better. The bottleneck is the word sound. Every candidate has to be listened to, the next training run waits on each listening round, and the automatic metrics I tried kept disagreeing with my ears.

Perceptual-Metrics on DRM HSE

A Tiny Speech Judge That Agrees With My Ears