Take audit: MsFHGUsUo_2BCFqiS8je0
2026-08-22T08:42:51.693Z — no tester given
0/13 judged · keyboard: Space play/pause, ↑/↓ focus, P pass, T Translation failure, F fail
Do a blind check (transcripts hidden)
Ground Truth
What the Tester actually said. Never pre-filled from an Engine — type what you hear.
Transcripts
mini_prod_baseline gpt-4o-mini-transcribe · degraded · en · bare
— MER accuracy pending Ground Truth · 908 ms · verdict: unjudged
老師,這是apple juice嗎?
mini_auto_bare_deg gpt-4o-mini-transcribe · degraded · auto · bare
— MER accuracy pending Ground Truth · 607 ms · verdict: unjudged
老師,這是apple juice嗎?
mini_auto_bare_clean gpt-4o-mini-transcribe · clean · auto · bare
— MER accuracy pending Ground Truth · 721 ms · verdict: unjudged
老師,這是apple juice嗎?
mini_auto_prompt_deg gpt-4o-mini-transcribe · degraded · auto · mixed prompt
— MER accuracy pending Ground Truth · 643 ms · verdict: unjudged
老師,這是apple juice嗎?
mini_auto_prompt_clean gpt-4o-mini-transcribe · clean · auto · mixed prompt
— MER accuracy pending Ground Truth · 731 ms · verdict: unjudged
老師,這是apple juice嗎?
mini_auto_prompt_64k gpt-4o-mini-transcribe · 64k · auto · mixed prompt
— MER accuracy pending Ground Truth · 1002 ms · verdict: unjudged
老師,這是apple juice嗎?
mini_zh_prompt_deg gpt-4o-mini-transcribe · degraded · zh · mixed prompt
— MER accuracy pending Ground Truth · 667 ms · verdict: unjudged
老師,這是apple juice嗎?
mini_zh_prompt_clean gpt-4o-mini-transcribe · clean · zh · mixed prompt
— MER accuracy pending Ground Truth · 903 ms · verdict: unjudged
老師,這是Apple juice嗎?
gpt_tw_en_bare_deg gpt-transcribe · degraded · zh-tw + en · bare
— MER accuracy pending Ground Truth · 603 ms · verdict: unjudged
老師,這是 Apple Juice 嗎?
gpt_tw_en_prompt_deg gpt-transcribe · degraded · zh-tw + en · mixed prompt
— MER accuracy pending Ground Truth · 631 ms · verdict: unjudged
老師,這是Apple Juice嗎?
gpt_tw_en_prompt_clean gpt-transcribe · clean · zh-tw + en · mixed prompt
— MER accuracy pending Ground Truth · 700 ms · verdict: unjudged
老師,這是 Apple Juice 嗎?
gpt_tw_en_prompt_64k gpt-transcribe · 64k · zh-tw + en · mixed prompt
— MER accuracy pending Ground Truth · 1124 ms · verdict: unjudged
老師,這是 Apple Juice 嗎?
gpt_tw_only_prompt_deg gpt-transcribe · degraded · zh-tw · mixed prompt
— MER accuracy pending Ground Truth · 733 ms · verdict: unjudged
老師,這是Apple Juice嗎?
Winning configuration
Record the variant and rationale to hand to qa-rails after reviewing this Take.