Over the weekend, I discovered a Whisper baseline that performed surprisingly well — so in today’s stream we focused on testing multiple Whisper models.

The big question: Is Whisper actually the key to unlocking better scores in this competition? We now have code ready to test and compare Whisper baselines against our existing work. Results are interesting, but you’ll have to catch the replay (or the next stream) to see if Whisper really is the game changer! 👀

📺 Replay : https://youtu.be/FfTc-GrzqLc

📺past streams: https://www.youtube.com/@koleshjr

🎮 Watch live: https://www.twitch.tv/koleshjr/schedule