fftfilter Namespace Reference

Classes
class	fftfilter_t
class	interface_t

Functions
unsigned int	irs_length (const MHAParser::mfloat_t &irs)
unsigned int	irs_validator (const MHAParser::mfloat_t &irs, const unsigned int &channels, const unsigned int &fragsize, const unsigned int &fftlen)

Function Documentation

irs_length()

unsigned int fftfilter::irs_length

(

const MHAParser::mfloat_t &

irs

)

Return the length of the longest vector in irs.

Parameters

irs	"Matrix" of floats parser variable

Returns

length of the longest vector in irs, or 1 if all vectors are empty

irs_validator()

unsigned int fftfilter::irs_validator	(	const MHAParser::mfloat_t &	irs,
		const unsigned int &	channels,
		const unsigned int &	fragsize,
		const unsigned int &	fftlen
	)

Validity checks. Throws Error if parameters are invalid: Number of channels must be > 0, fftlen must be >= fragsize. The number of rows in irs has to match the number of channels, or has to be exactly 1. None of the row vectors may be empty. The longest supported impulse response is (fftlen - fragsize + 1). Impulse responses longer than (fftlen - fragsize + 1) would cause temporal aliasing.

Parameters

irs	The matrix containing the impulse responses (one response per channel, or the same response for every channels) as set by the parser.
channels	The number of prepared audio channels for this MHA plugin.
fragsize	The block size (samples per channel) for waveform audio data
fftlen	FFT length used for filtering.

Returns

the length of the longest impulse response vector in irs.