OptionalaudioOptionalbeamOptionalbestOptionalfastExecution path: true (default) runs the optimized graphs
(the graph's fast path) for temperature-0 decoding; false forces the
faithful oracle graph. Sampling (temperature > 0) always uses the
faithful path.
Optionallanguagenull → detect.
OptionallengthOptionalmaxOptionalobserverParity instrumentation: the decode window's encoder output and each step's PRE-filter last-position logits [n_group, V] f32 (borrowed — valid only during the callback).
OptionalpatienceOptionalprefixOptionalpromptOptionalsampleOptionalsuppressOptionalsuppressToken ids to suppress; "-1" = the tokenizer's non-speech set (default).
OptionaltaskOptionaltemperatureOptionalwithout
Lab (whisper.cpp
-ac): encode only the firstaudioCtxof the 1500 encoder positions (2·audioCtx mel frames). Cheaper for short clips but NOT the trained context — measured hallucinated repeats below ~1024 on the fixture clips. null = full context. Fast path only.