Page MenuHome
Differential D2716 Diff 9153 extern/audaspace/include/fx/FFTConvolver.h

Changeset View

Standalone View

View Options

extern/audaspace/include/fx/FFTConvolver.h

  • This file was added.
/*******************************************************************************
* Copyright 2015-2016 Juan Francisco Crespo Galán
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
******************************************************************************/
#pragma once
/**
* @file FFTConvolver.h
* @ingroup fx
* The FFTConvolver class.
*/
#include "IReader.h"
#include "ISound.h"
#include "util/FFTPlan.h"
#include <memory>
#include <vector>
AUD_NAMESPACE_BEGIN
/**
* This class allows to easily convolve a sound using the Fourier transform.
*/
class AUD_API FFTConvolver
{
private:
/**
* A shared pointer to an FFT plan.
*/
std::shared_ptr<FFTPlan> m_plan;
/**
* The FFT size, must be at least M+L-1.
*/
int m_N;
/**
* The length of the impulse response.
*/
int m_M;
/**
* The max length of the input slices.
*/
int m_L;
/**
* The real length of the internal buffer in fftwf_complex elements.
*/
int m_realBufLen;
/**
* The internal buffer for the FFTS.
*/
std::complex<sample_t>* m_inBuffer;
/**
* A shift buffer for the FDL method
*/
sample_t* m_shiftBuffer;
/**
* A buffer to store the extra data obtained after each partial convolution.
*/
float* m_tail;
/**
* The provided impulse response.
*/
std::shared_ptr<std::vector<std::complex<sample_t>>> m_irBuffer;
/**
* If the tail is being read, this marks the current position.
*/
int m_tailPos;
// delete copy constructor and operator=
FFTConvolver(const FFTConvolver&) = delete;
FFTConvolver& operator=(const FFTConvolver&) = delete;
public:
/**
* Creates a new FFTConvolver.
* \param ir A shared pointer to a vector with the impulse response data in the frequency domain (see ImpulseResponse class for an easy way to obtain it).
* \param plan A shared pointer to and FFT plan.
*/
FFTConvolver(std::shared_ptr<std::vector<std::complex<sample_t>>> ir, std::shared_ptr<FFTPlan> plan);
virtual ~FFTConvolver();
/**
* Convolves the data that is provided with the inpulse response.
* \param[in] inBuffer A buffer with the input data to be convolved.
* \param[in] outBuffer A pointer to the buffer in which the convolution result will be written.
* \param[in,out] length The number of samples to be convolved (the length of both the inBuffer and the outBuffer).
* The convolution output should be larger than the input, but since this class uses the overlap
* add method, the extra length will be saved internally.
* It must be equal or lower than N/2 (N=size of the FFTPlan) or the call will fail, setting this variable to 0 since no data would be
* written in the outBuffer.
*/
void getNext(const sample_t* inBuffer, sample_t* outBuffer, int& length);
/**
* Convolves the data that is provided with the inpulse response.
* \param[in] inBuffer A buffer with the input data to be convolved.
* \param[in] outBuffer A pointer to the buffer in which the convolution result will be written.
* \param[in,out] length The number of samples to be convolved (the length of both the inBuffer and the outBuffer).
* The convolution output should be larger than the input, but since this class uses the overlap
* add method, the extra length will be saved internally.
* It must be equal or lower than N/2 (N=size of the FFTPlan) or the call will fail, setting this variable to 0 since no data would be
* written in the outBuffer.
* \param[in] transformedData A pointer to a buffer in which the Fourier transform of the input will be written.
*/
void getNext(const sample_t* inBuffer, sample_t* outBuffer, int& length, fftwf_complex* transformedData);
/**
* Convolves the data that is provided with the inpulse response.
* \param[in] inBuffer A buffer with the input data to be convolved. Its length must be N/2 + 1
* \param[in] outBuffer A pointer to the buffer in which the convolution result will be written.
* \param[in,out] length The number of samples to be convolved and the length of the outBuffer.
* The convolution output should be larger than the input, but since this class uses the overlap
* add method, the extra length will be saved internally.
* It must be equal or lower than N/2 (N=size of the FFTPlan) or the call will fail and set the value of length to 0 since no data would be
* written in the outBuffer.
*/
void getNext(const fftwf_complex* inBuffer, sample_t* outBuffer, int& length);
/**
* Gets the internally stored extra data which is result of the convolution.
* \param[in,out] length The count of samples that should be read. Shall
* contain the real count of samples after reading, in case
* there were only fewer samples available.
* A smaller value also indicates the end of the data.
* \param[out] eos End of stream, whether the end is reached or not.
* \param[in] buffer The pointer to the buffer to read into.
*/
void getTail(int& length, bool& eos, sample_t* buffer);
/**
* Resets the internally stored data so a new convolution can be started.
*/
void clear();
/**
* Calculates the Inverse Fast Fourier Transform of the input array.
* \param[in] inBuffer A buffer with the input data to be transformed. Its length must be N/2 + 1
* \param[in] outBuffer A pointer to the buffer in which the transform result will be written.
* \param[in,out] length The number of samples to be transformed and the length of the outBuffer.
* It must be equal or lower than N, but tipically N/2 should be used (N=size of the FFTPlan) or the call will fail and the value
* of length will be setted to 0, since no data would be written in the outBuffer.
*/
void IFFT_FDL(const fftwf_complex* inBuffer, sample_t* outBuffer, int& length);
/**
* Multiplicates a frequency domain input by the impulse response and accumulates the result to a buffer.
* \param[in] inBuffer A buffer of complex numbers, samples in the frequency domain, that will be multiplied by the impulse response. Its length must be N/2 + 1
* \param[in] accBuffer A pointer to the buffer into which the result of the multiplication will be summed. Its length must be N/2 + 1
*/
void getNextFDL(const std::complex<sample_t>* inBuffer, std::complex<sample_t>* accBuffer);
/**
* Transforms an input array of real data to the frequency domain and multiplies it by the impulse response. The result is accumulated to a buffer.
* \param[in] inBuffer A buffer of real numbers, samples in the time domain, that will be multiplied by the impulse response.
* \param[in] accBuffer A pointer to the buffer into which the result of the multiplication will be summed. Its length must be N/2 + 1.
* \param[in,out] length The number of samples to be transformed and the length of the inBuffer.
* It must be equal or lower than N/2 (N=size of the FFTPlan) or the call will fail and the value
* of length will be setted to 0, since no data would be written in the outBuffer.
* \param[in] transformedData A pointer to a buffer in which the Fourier transform of the input will be written.
*/
void getNextFDL(const sample_t* inBuffer, std::complex<sample_t>* accBuffer, int& length, fftwf_complex* transformedData);
/**
* Changes the impulse response and resets the FFTConvolver.
* \param ir A shared pointer to a vector with the data of the impulse response in the frequency domain.
*/
void setImpulseResponse(std::shared_ptr<std::vector<std::complex<sample_t>>> ir);
/**
* Retrieves the current impulse response being used.
* \return The current impulse response.
*/
std::shared_ptr<std::vector<std::complex<sample_t>>> getImpulseResponse();
};
AUD_NAMESPACE_END