Should resample based on requirements of underlying SnowboyDetect instance #7
Description
SnowboyDetect offers methods to request the required audio data properties for its correct working. At the node level, there this should be taken into account.
Would be nice if we can then also input an arbitrary sample rate/number of channels/bit depth/endianness and configure the node so that it correctly transforms the data from the input topic to data that the underlying SnowboyDetect node can use.