Streaming phase-gradient vocoder with a strike-anchored time map. More...

#include <StudioVocoder.h>

Classes
struct	Params
	The control-published parameter set (adopted as ONE unit). More...

Public Member Functions
bool	prepare (double sampleRate, int numChannels, int fftSize, bool resampleCompensation)
	Allocates every buffer the stream will use.

void	reset () noexcept
	Clears all signal state and adopts the latest parameters (jump, no glide). Stream owner only.

void	publishParams (const Params &p) noexcept
	Publishes a whole parameter set (one control thread only).

int	samplesToNextHop () const noexcept

void	pushInput (int ch, const T *src, int count) noexcept
	Writes `count` (<= samplesToNextHop(), or any count when it is 0... see commitInput) samples of channel `ch` at the input head.

void	commitInput (int count, int numActiveChannels) noexcept
	Advances the input head by `count` and runs at most one frame.

void	steerTimeline (double nominalCentre) noexcept
	Owner-supplied nominal analysis centre for the NEXT frame (a resampling owner keeps its reader and the timeline in step with it).

void	setAnchorLeadLimit (double samples) noexcept
	Caps how far a strike anchor may move the analysis ahead of the nominal timeline, in input samples (owner with a fixed latency budget; unlimited by default).

double	activeRatio () const noexcept

double	nextNominalCentre () const noexcept

double	nextStreamCentre () const noexcept

double	streamPositionOf (double x) const noexcept

const T *	olaData (int ch) const noexcept

int64_t	olaMask () const noexcept

int	olaSize () const noexcept

int64_t	writeHead () const noexcept

int	fftSize () const noexcept

int	synthHop () const noexcept

int	lookahead () const noexcept

int64_t	inputCount () const noexcept

Detailed Description

template<FloatType T>
class dspark::detail::StudioVocoder< T >

Streaming phase-gradient vocoder with a strike-anchored time map.

Template Parameters

T	Sample type (float or double).

Definition at line 98 of file StudioVocoder.h.

Member Function Documentation

◆ activeRatio()

template<FloatType T>

double dspark::detail::StudioVocoder< T >::activeRatio ( ) const

inlinenoexcept

Returns: The glided active ratio (stretch factor, stream per input).

Definition at line 326 of file StudioVocoder.h.

◆ commitInput()

template<FloatType T>

void dspark::detail::StudioVocoder< T >::commitInput	(	int	count,
		int	numActiveChannels
	)

inlinenoexcept

Advances the input head by count and runs at most one frame.

The onset detector consumes the new input first, so a strike it finds is planned before the frame that would reach it is analysed.

Definition at line 297 of file StudioVocoder.h.

◆ fftSize()

template<FloatType T>

int dspark::detail::StudioVocoder< T >::fftSize ( ) const

inlinenoexcept

Definition at line 348 of file StudioVocoder.h.

◆ inputCount()

template<FloatType T>

int64_t dspark::detail::StudioVocoder< T >::inputCount ( ) const

inlinenoexcept

Returns: Input samples committed since reset().

Definition at line 353 of file StudioVocoder.h.

◆ lookahead()

template<FloatType T>

int dspark::detail::StudioVocoder< T >::lookahead ( ) const

inlinenoexcept

Returns: How far (samples) the analysis trails the newest input.

Definition at line 351 of file StudioVocoder.h.

◆ nextNominalCentre()

template<FloatType T>

double dspark::detail::StudioVocoder< T >::nextNominalCentre ( ) const

inlinenoexcept

Returns: Nominal analysis centre (input coordinate) of the next frame.

Definition at line 329 of file StudioVocoder.h.

◆ nextStreamCentre()

template<FloatType T>

double dspark::detail::StudioVocoder< T >::nextStreamCentre ( ) const

inlinenoexcept

Returns: Stream centre of the next frame.

Definition at line 332 of file StudioVocoder.h.

◆ olaData()

template<FloatType T>

const T * dspark::detail::StudioVocoder< T >::olaData ( int ch ) const

inlinenoexcept

Definition at line 343 of file StudioVocoder.h.

◆ olaMask()

template<FloatType T>

int64_t dspark::detail::StudioVocoder< T >::olaMask ( ) const

inlinenoexcept

Definition at line 344 of file StudioVocoder.h.

◆ olaSize()

template<FloatType T>

int dspark::detail::StudioVocoder< T >::olaSize ( ) const

inlinenoexcept

Definition at line 345 of file StudioVocoder.h.

◆ prepare()

template<FloatType T>

bool dspark::detail::StudioVocoder< T >::prepare	(	double	sampleRate,
		int	numChannels,
		int	fftSize,
		bool	resampleCompensation
	)

inline

Allocates every buffer the stream will use.

Parameters

sampleRate	Sample rate in Hz.
numChannels	Channel count (>= 1).
fftSize	Frame size, power of two in [256, 1 << 20].
resampleCompensation	True for owners that resample the stream by the ratio afterwards (pitch shifting): enables the anti-alias taper and the formant pre-warp.

Returns: false (nothing touched) on invalid arguments.

Definition at line 129 of file StudioVocoder.h.

◆ publishParams()

template<FloatType T>

void dspark::detail::StudioVocoder< T >::publishParams ( const Params & p )

inlinenoexcept

Publishes a whole parameter set (one control thread only).

Definition at line 260 of file StudioVocoder.h.

◆ pushInput()

template<FloatType T>

void dspark::detail::StudioVocoder< T >::pushInput	(	int	ch,
		const T *	src,
		int	count
	)

inlinenoexcept

Writes count (<= samplesToNextHop(), or any count when it is 0... see commitInput) samples of channel ch at the input head.

Definition at line 283 of file StudioVocoder.h.

◆ reset()

template<FloatType T>

void dspark::detail::StudioVocoder< T >::reset ( )

inlinenoexcept

Clears all signal state and adopts the latest parameters (jump, no glide). Stream owner only.

Definition at line 212 of file StudioVocoder.h.

◆ samplesToNextHop()

template<FloatType T>

int dspark::detail::StudioVocoder< T >::samplesToNextHop ( ) const

inlinenoexcept

Returns: Input samples to push before the next frame can run (0 = the next commitInput() runs it).

Definition at line 275 of file StudioVocoder.h.

◆ setAnchorLeadLimit()

template<FloatType T>

void dspark::detail::StudioVocoder< T >::setAnchorLeadLimit ( double samples )

inlinenoexcept

Caps how far a strike anchor may move the analysis ahead of the nominal timeline, in input samples (owner with a fixed latency budget; unlimited by default).

A lock of half-width h runs h (1/r - 1) ahead of the timeline as it starts below ratio 1, and h (1 - 1/r) ahead as it ends above it; the cap narrows the lock instead of letting a frame need input the owner's latency does not cover.

Definition at line 323 of file StudioVocoder.h.

◆ steerTimeline()

template<FloatType T>

void dspark::detail::StudioVocoder< T >::steerTimeline ( double nominalCentre )

inlinenoexcept

Owner-supplied nominal analysis centre for the NEXT frame (a resampling owner keeps its reader and the timeline in step with it).

Definition at line 307 of file StudioVocoder.h.

◆ streamPositionOf()

template<FloatType T>

double dspark::detail::StudioVocoder< T >::streamPositionOf ( double x ) const

inlinenoexcept

Returns: Stream position of input x on the timeline in force at reset.

Definition at line 338 of file StudioVocoder.h.

◆ synthHop()

template<FloatType T>

int dspark::detail::StudioVocoder< T >::synthHop ( ) const

inlinenoexcept

Definition at line 349 of file StudioVocoder.h.

◆ writeHead()

template<FloatType T>

int64_t dspark::detail::StudioVocoder< T >::writeHead ( ) const

inlinenoexcept

Returns: End of the completed overlap-add (advances Rs per frame).

Definition at line 347 of file StudioVocoder.h.

The documentation for this class was generated from the following file:

StudioVocoder.h

Classes

Public Member Functions

Detailed Description

Member Function Documentation

◆ activeRatio()

◆ commitInput()

◆ fftSize()

◆ inputCount()

◆ lookahead()

◆ nextNominalCentre()

◆ nextStreamCentre()

◆ olaData()

◆ olaMask()

◆ olaSize()

◆ prepare()

◆ publishParams()

◆ pushInput()

◆ reset()

◆ samplesToNextHop()

◆ setAnchorLeadLimit()

◆ steerTimeline()

◆ streamPositionOf()

◆ synthHop()

◆ writeHead()