MFPKEY_WMAAECMA_FEATR_FRAME_SIZE Property

Specifies the audio frame size used by the Voice Capture DSP.

Constant for IPropertyBag

Available only by using IPropertyStore.

Data Type

VT_I4

Default Value

0

Applies To

• Voice Capture DSP

Remarks

The acoustic echo cancellation (AEC) algorithm processes PCM audio samples one frame at a time. The value of this property is the size of the audio frame, in samples. Before setting this property, you must set the MFPKEY_WMAAECMA_FEATURE_MODE property to VARIANT_TRUE.

The Voice Capture DSP supports the following frame sizes:

• 80
• 128
• 160
• 240
• 256
• 320

If the value of this property is zero, the DSP selects the frame size based on the system mode and the output format.

For the best performance, however, it is recommended that applications use the default value. If the processing mode is microphone array only, the default value is 320 samples. For all other processing modes, the default value is 160 samples. For more information about the processing modes of the Voice Capture DSP, see MFPKEY_WMAAECMA_SYSTEM_MODE.

After the first call to IMediaObject::AllocateStreamingResources or IMediaObject::ProcessOutput, you can read this property to get the actual frame size in use, even when MFPKEY_WMAAECMA_FEATURE_MODE is false.

Requirements

Requirement	Value
Minimum supported client	Windows Vista [desktop apps only]
Minimum supported server	Windows Server 2008 [desktop apps only]
Header	Wmcodecdsp.h

See also

Media Foundation Properties

Voice Capture DSP