ResamplePolyphase on multichannel PCM

[svilst]

If I have multichannel interleaved audio it is clear that I should deinterleave the samples and apply ippsResamplePolyphase on each channel.

But is it OK to use ippsResamplePolyphaseInitAlloc just once to obtain a resampling state structure (pSpec) and then call ippsResamplePolyphase with this pSpec for the deinterleaved samples of each channel ?

Or I should call ippsResamplePolyphaseInitAlloc for each channel and obtain resampling state structures (pSpec) for each channel?

Obviously the answer depends on whether the resampling state (pSpec) stores information about the already processed samples or not but I could not find it in the documentation.

Thanks,
Svilen

[Chao_Y_Intel · Accepted solution]

Hi Svilen,

If they have sampe paramters ( in, out rate, filter length), the structures can be reused.

Some Intel IPP functions use special structures to store function-specific (context) information.

Two different kinds of structures are used: specification structures that are not modified during

function operation -they have the suffix Spec in their names, and state structures that are

modified during operation - they have the suffix State in their names.

resampling function (use pSpec structures) are not modified during function operation.

Thanks,

Chao

[svilst]

Thank you for the clarification.