TinyAya v0.3 open speech-to-speech model adds text supervision and 36-layer LoRA

irombie · x · 2026-07-23

TinyAya v0.3 is an open Turkish↔Hindi speech-to-speech translation model built on a Cohere2 backbone with LoRA fine-tuning and a frozen Moshi depth decoder.

Key details from the thread:

The post says all recipes, configs, W&B runs, and postmortems are open.

Original post →

More from Research

Research channel →