Take audit: fPZvQa2JjYA1vyaCX28su

2026-08-22T09:35:34.360Z — 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 · 966 ms · verdict: unjudged

Let's choose Welsh in

mini_auto_bare_deg gpt-4o-mini-transcribe · degraded · auto · bare — MER accuracy pending Ground Truth · 597 ms · verdict: unjudged

Let's choose Welsh.

mini_auto_bare_clean gpt-4o-mini-transcribe · clean · auto · bare — MER accuracy pending Ground Truth · 615 ms · verdict: unjudged

Let's Choose我要選這個

mini_auto_prompt_deg gpt-4o-mini-transcribe · degraded · auto · mixed prompt — MER accuracy pending Ground Truth · 633 ms · verdict: unjudged

Let's choose well,選擇吧

mini_auto_prompt_clean gpt-4o-mini-transcribe · clean · auto · mixed prompt — MER accuracy pending Ground Truth · 802 ms · verdict: unjudged

Let's choose 我要選這個

mini_auto_prompt_64k gpt-4o-mini-transcribe · 64k · auto · mixed prompt — MER accuracy pending Ground Truth · 651 ms · verdict: unjudged

Let's choose我要選這個

mini_zh_prompt_deg gpt-4o-mini-transcribe · degraded · zh · mixed prompt — MER accuracy pending Ground Truth · 632 ms · verdict: unjudged

Let's choose well選擇。

mini_zh_prompt_clean gpt-4o-mini-transcribe · clean · zh · mixed prompt — MER accuracy pending Ground Truth · 705 ms · verdict: unjudged

Let's choose 我要選這個

gpt_tw_en_bare_deg gpt-transcribe · degraded · zh-tw + en · bare — MER accuracy pending Ground Truth · 549 ms · verdict: unjudged

Let's choose，我要選這個。

gpt_tw_en_prompt_deg gpt-transcribe · degraded · zh-tw + en · mixed prompt — MER accuracy pending Ground Truth · 738 ms · verdict: unjudged

That's choose，我要選這個。

gpt_tw_en_prompt_clean gpt-transcribe · clean · zh-tw + en · mixed prompt — MER accuracy pending Ground Truth · 711 ms · verdict: unjudged

That's choose 我要選這個。

gpt_tw_en_prompt_64k gpt-transcribe · 64k · zh-tw + en · mixed prompt — MER accuracy pending Ground Truth · 657 ms · verdict: unjudged

That's choose 我要選這個。

gpt_tw_only_prompt_deg gpt-transcribe · degraded · zh-tw · mixed prompt — MER accuracy pending Ground Truth · 732 ms · verdict: unjudged

That's choose，我要選這個。

Winning configuration

Record the variant and rationale to hand to qa-rails after reviewing this Take.