@_david_7060

videoMD · tier B

$_david_$ · 438 followers · 2026-07-06 14:45 · 41s

Audio track (133.6 KB), what ASR actually heard

16 kHz mono Opus. Ogg Opus does not play in Safari; Chrome and Firefox are fine.

eagle · fetched 2026-07-26 05:44 · 1.5 MB downloaded · 4 keyframes

plays

193

likes

comments

Topic

food_cookingconf 0.80positive

Fixed taxonomy enforced as a schema enum. Prose instructions did not hold, and label drift makes week-over-week comparison meaningless.

Entities

Subway

Keyphrases

mic frappé

gpt-4.1-nano · prompt v4 · $0.000092 · 2026-07-26 08:50

Caption

Author-written, as posted.

No caption. True for 83.6% of the corpus.

Speech (ASR)

speechfr 90%

faster-whisper large-v3-turbo behind Silero VAD, then a repetition filter.

Subway, mic frappé, Skibidi Mbappé

rtx4090-24-02 · 34 chars · 8.7s speech of 41.1s

On-screen text (OCR)

easyocr

EasyOCR for Latin script, macOS Vision for Cyrillic-suspect frames.

M B

rtx4090-24-02 · 3 chars · 2 frames