AudioTranscribeWhisperConfiguration
Field |
Type |
Repeated |
Description |
id |
|||
output_stream_id |
|||
step_ms |
|||
keep_ms |
|||
max_tokens |
|||
speed_up |
|||
no_fallback |
|||
num_threads |
|||
use_gpu |
|||
language |
|||
model |
|||
translate |
Translate (to English) |
||
tiny_diarize |
|||
initial_prompt |
|||
suppress_non_speech_tokens |
|||
sampling_strategy |
|||
best_of |
|||
beam_size |
|||
context_prompt |
Whether to supply a prompt consisting of the tokens recognised in the previous chunk |
||
initial_temperature |
Initial temperature (unset: use library default, currently 0.0) |
||
temperature_increment |
Temperature increment on fallback (unset: use library default, currently 0.2). To disable fallback set this explicitly to 0.0 |
||
entropy_threshold |
|||
log_prob_threshold |
|||
no_speech_threshold |
Enums defined in this message
Name |
Number |
Description |
WHISPER_SAMPLING_GREEDY |
0 |
|
WHISPER_SAMPLING_BEAM_SEARCH |
1 |