Take audit: lG_11cYyJVoLGy_osrQDL
2026-08-22T09:34:07.868Z — ray
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 · 637 ms · verdict: unjudged
This is for you, 謝謝,謝老師。
mini_auto_bare_deg gpt-4o-mini-transcribe · degraded · auto · bare
— MER accuracy pending Ground Truth · 647 ms · verdict: unjudged
This is for you, 謝老師
mini_auto_bare_clean gpt-4o-mini-transcribe · clean · auto · bare
— MER accuracy pending Ground Truth · 612 ms · verdict: unjudged
This is for you,寫寫寫老師。
mini_auto_prompt_deg gpt-4o-mini-transcribe · degraded · auto · mixed prompt
— MER accuracy pending Ground Truth · 660 ms · verdict: unjudged
This is for you, 謝謝,謝老師
mini_auto_prompt_clean gpt-4o-mini-transcribe · clean · auto · mixed prompt
— MER accuracy pending Ground Truth · 723 ms · verdict: unjudged
This is for you,寫老師。
mini_auto_prompt_64k gpt-4o-mini-transcribe · 64k · auto · mixed prompt
— MER accuracy pending Ground Truth · 854 ms · verdict: unjudged
This is for you, CS, 沈老師
mini_zh_prompt_deg gpt-4o-mini-transcribe · degraded · zh · mixed prompt
— MER accuracy pending Ground Truth · 730 ms · verdict: unjudged
This is for you, 謝謝,謝老師。
mini_zh_prompt_clean gpt-4o-mini-transcribe · clean · zh · mixed prompt
— MER accuracy pending Ground Truth · 729 ms · verdict: unjudged
This is for you,寫老師。
gpt_tw_en_bare_deg gpt-transcribe · degraded · zh-tw + en · bare
— MER accuracy pending Ground Truth · 786 ms · verdict: unjudged
This is for you, 謝謝老師.
gpt_tw_en_prompt_deg gpt-transcribe · degraded · zh-tw + en · mixed prompt
— MER accuracy pending Ground Truth · 771 ms · verdict: unjudged
This is for you, 謝謝老師。
gpt_tw_en_prompt_clean gpt-transcribe · clean · zh-tw + en · mixed prompt
— MER accuracy pending Ground Truth · 779 ms · verdict: unjudged
This is for you. 謝謝老師。
gpt_tw_en_prompt_64k gpt-transcribe · 64k · zh-tw + en · mixed prompt
— MER accuracy pending Ground Truth · 761 ms · verdict: unjudged
This is for you, 謝謝老師.
gpt_tw_only_prompt_deg gpt-transcribe · degraded · zh-tw · mixed prompt
— MER accuracy pending Ground Truth · 812 ms · verdict: unjudged
This is for you, 謝謝老師。
Winning configuration
Record the variant and rationale to hand to qa-rails after reviewing this Take.