Megadose AI progress, ranked and analyzed.

Building a Production Greek-English Speech Recognizer

· HF Daily Papers ·
The team’s single-model runs could not satisfy all nine production gates at once.

Sophea was tested across Greek and English error rates, language ID, and hallucinations on non-speech audio. The paper says Greek noisy-traffic performance needed much longer dense domain exposure than English language-ID preservation could tolerate. A three-model ROVER ensemble passed all nine gates and cut overlapping-speech WER from 53.35% to 37.87%. The authors release methodology and results, but no model weights or training data. HF Daily Papers' note

score 4

Categories: Research