Edit

Share via

AEC_VAD_MODE enumeration (wmcodecdsp.h)

Specifies the type of voice activity detection (VAD) for the voice capture DSP. This enumeration is used with the MFPKEY_WMAAECMA_FEATR_VAD property.

Syntax

typedef enum AEC_VAD_MODE {
  AEC_VAD_DISABLED = 0,
  AEC_VAD_NORMAL,
  AEC_VAD_FOR_AGC,
  AEC_VAD_FOR_SILENCE_SUPPRESSION
} ;

Constants

 
AEC_VAD_DISABLED
Value: 0
Voice activity detection is disabled.
AEC_VAD_NORMAL
General purpose VAD. This setting attempts to balance the false detection rate and the missed detection rate. The output can have the following values.























ValueDescription
0Non-speech.
1Voiced speech.
2Unvoiced speech.
3Mixture of voiced and unvoiced speech.
AEC_VAD_FOR_AGC
Voice activity detection for automatic gain control and noise suppression. In this mode, the VAD threshold is higher than the normal mode, to reduce the false detection rate. The output distinguishes voiced speech from other sounds (non-speech or unvoiced speech). The output can have the following values.















ValueDescription
0Non-speech or unvoiced speech.
1Voiced speech.
AEC_VAD_FOR_SILENCE_SUPPRESSION
Voice activity detection for silence suppression. In this mode, the output distinguishes speech (voiced or unvoiced) from non-speech. The output can have the following values.















ValueDescription
0Non-speech.
1Voiced or unvoiced speech.

Requirements

Requirement Value
Minimum supported client Windows XP [desktop apps only]
Minimum supported server Windows Server 2003 [desktop apps only]
Header wmcodecdsp.h

See also

Media Foundation Enumerations

Voice Capture