A bit surreal to see Open ASR Leaderboard be cited by ~35 papers - this started as a side project with a handful of models and now we're pretty much the industry standard for both closed and open speech recognition models! Thanks a lot for the trust and support 🙏