Voice Activity Detection Node

V

AI/ML/ONNX/Audio

Voice Activity Detection

Detect speech segments in audio. Download Silero VAD model from: https://github.com/snakers4/silero-vad/raw/master/src/silero_vad/data/silero_vad.onnx

onnx_vadonnx

Inputs6

Outputs3

Security exposure10/10

Packageonnx

Ratings

Scores range from 0 to 10. Higher values mean more impact, exposure, or operational weight.

No score metadata has been set for this node yet.

Input Pins

6

Input

Execution

exec_in

Initiate Execution

Model

Struct

model

ONNX VAD Model

SchemaNodeOnnxSessionNodeOnnxSessionNodeOnnxSession1 fields

session_refstringrequired

Cache ID for Session

Schema enforced

Audio

Struct

audio

Input audio data

SchemaAudioDataAudioDataAudioData4 fields

sample_rateinteger:uint32required

Sample rate in Hz

format uint32min 0

channelsinteger:uint16required

Number of channels (1 = mono, 2 = stereo)

format uint16min 0max 65535

samplesArray<number:float>required

Audio samples (normalized to -1.0 to 1.0)

itemsnumber:floatarray item

format float

duration_secsnumber:floatrequired

Duration in seconds

format float

Threshold

Float

threshold

Speech probability threshold

Default 0.5

Min Speech

Integer

min_speech_ms

Minimum speech duration (ms)

Default 250

Min Silence

Integer

min_silence_ms

Minimum silence duration (ms)

Default 100

Output Pins

3

Output

Execution

exec_out

Done

Result

Struct

result

VAD result

SchemaVadResultVadResultVadResult3 fields

segmentsArray<SpeechSegment>required

Speech segments detected

itemsSpeechSegmentarray item

startnumber:floatrequired

Start time in seconds

format float

endnumber:floatrequired

End time in seconds

format float

confidencenumber:floatrequired

Average confidence

format float

probabilitiesArray<number:float>required

Frame-level speech probabilities

itemsnumber:floatarray item

format float

frame_durationnumber:floatrequired

Frame duration in seconds

format float

Segments

Generic

segments

Speech segments

Node Info

Internal name

onnx_vad

Category

AI/ML/ONNX/Audio

Version

1