AudioTranscribeWhisperConfiguration

Field

Type

Repeated

Description

id

MediaNodeId

output_stream_id

uint32

step_ms

OptionalInt

length_ms

OptionalInt

keep_ms

OptionalInt

max_tokens

OptionalInt

speed_up

OptionalBool

no_fallback

OptionalBool

num_threads

OptionalInt

use_gpu

OptionalBool

language

string

model

string

translate

bool

Translate (to English)

tiny_diarize

bool

initial_prompt

string

suppress_non_speech_tokens

OptionalBool

sampling_strategy

AudioTranscribeWhisperConfiguration.SamplingStrategy

best_of

OptionalInt

beam_size

OptionalInt

context_prompt

OptionalBool

Whether to supply a prompt consisting of the tokens recognised in the previous chunk

Enums defined in this message

Name

Number

Description

WHISPER_SAMPLING_GREEDY

0

WHISPER_SAMPLING_BEAM_SEARCH

1

Member of

Message

Description

AudioTranscribeWhisperMessage