Skip to content

Class lra_vec

class lra_vec<T> dsp

template <typename T>
struct lra_vec : public univector<T> { … }

Sorted buffer of short-term mean-square energies used to compute the Loudness Range (LRA) per EBU R128 / ITU-R BS.1770.

Samples below -70 LUFS are rejected on push; the rest are kept in ascending order so that percentile lookups are O(log n). LRA is the difference between the 95 % and 10 % percentiles of the loudness distribution above a relative gate set 20 LU below the ungated mean.

constructor lra_vec<T>() 

lra_vec()

Fully qualified name: kfr::lra_vec<T>::lra_vec<T>()

Declared at dsp/ebu.hpp:194

function push(T) 

void push(T mean_square)

Insert a short-term mean-square energy sample, keeping the buffer sorted.

Samples whose loudness is below -70 LUFS are rejected.

Parameters
mean_square Short-term mean-square energy for one 3 s window.

Fully qualified name: kfr::lra_vec<T>::push(T)

Declared at dsp/ebu.hpp:201

function reset() 

void reset()

Clear the buffer and invalidate the cached result.

Fully qualified name: kfr::lra_vec<T>::reset()

Declared at dsp/ebu.hpp:212

function get(T &, T &) 

void get(T& low, T& high) const

Return the LRA percentile loudness bounds.

Parameters
low Out: 10 % percentile loudness in LUFS.
high Out: 95 % percentile loudness in LUFS.

Note

LRA = high - low.

Fully qualified name: kfr::lra_vec<T>::get(T &, T &)

Declared at dsp/ebu.hpp:223

Fully qualified name: kfr::lra_vec<T>

Declared at dsp/ebu.hpp:160

This file was generated by cxxdox, a C++ documentation generator based on MkDocs Material and libclang.