SRC (libsamplerate) resampler. More...
#include <ResamplerSrc.h>
Public Types | |
| enum class | Quality { Linear , ZeroOrderHold , SincFastest , SincMedium , SincBest } |
| SRC resampler quality. More... | |
Public Member Functions | |
| ResamplerSrc (Quality quality=Quality::SincMedium) | |
| ~ResamplerSrc () override | |
| auto | quality () const noexcept -> Quality |
 Public Member Functions inherited from Aulib::Resampler | |
| Resampler () | |
| Constructs an audio resampler. More... | |
| virtual | ~Resampler () |
| Resampler (const Resampler &)=delete | |
| auto | operator= (const Resampler &) -> Resampler &=delete |
| void | setDecoder (std::shared_ptr< Decoder > decoder) |
| Sets the decoder that is to be used as source. More... | |
| auto | setSpec (int dstRate, int channels, int chunkSize) -> int |
| Sets the target sample rate, channels and chuck size. More... | |
| auto | currentRate () const -> int |
| auto | currentChannels () const -> int |
| auto | currentChunkSize () const -> int |
| auto | resample (float dst[], int dstLen) -> int |
| Fills an output buffer with resampled audio samples. More... | |
| void | discardPendingSamples () |
| Discards any samples that have not yet been retrieved with resample(). More... | |
Protected Member Functions | |
| void | doResampling (float dst[], const float src[], int &dstLen, int &srcLen) override |
| auto | adjustForOutputSpec (int dstRate, int srcRate, int channels) -> int override |
| Change sample rate and amount of channels. More... | |
| void | doDiscardPendingSamples () override |
| Discard any internally held samples. More... | |
| virtual auto | adjustForOutputSpec (int dstRate, int srcRate, int channels) -> int=0 |
| Change sample rate and amount of channels. More... | |
| virtual void | doResampling (float dst[], const float src[], int &dstLen, int &srcLen)=0 |
| virtual void | doDiscardPendingSamples ()=0 |
| Discard any internally held samples. More... | |
Detailed Description
SRC (libsamplerate) resampler.
Member Enumeration Documentation
◆ Quality
|
strong |
SRC resampler quality.
These represent the five SRC resampling methods, with "Linear" and "ZeroOrderHold" being the fastest but lowest quality, and "SincBest" the slowest but highest quality.
From SRC's API documentation:
- SincBest This is a bandlimited interpolator derived from the mathematical sinc function and this is the highest quality sinc based converter, providing a worst case Signal-to-Noise Ratio (SNR) of 97 decibels (dB) at a bandwidth of 97%. All three Sinc* converters are based on the techniques of Julius O. Smith although this code was developed independantly.
- SincMedium This is another bandlimited interpolator much like the previous one. It has an SNR of 97dB and a bandwidth of 90%. The speed of the conversion is much faster than the previous one.
- SincFastest This is the fastest bandlimited interpolator and has an SNR of 97dB and a bandwidth of 80%.
- ZeroOrderHold A Zero Order Hold converter (interpolated value is equal to the last value). The quality is poor but the conversion speed is blindlingly fast.
- Linear A linear converter. Again the quality is poor, but the conversion speed is blindingly fast.
| Enumerator | |
|---|---|
| Linear | |
| ZeroOrderHold | |
| SincFastest | |
| SincMedium | |
| SincBest | |
Constructor & Destructor Documentation
◆ ResamplerSrc()
|
explicit |
- Parameters
-
quality Resampling quality. Note that the quality can not be changed later on.
◆ ~ResamplerSrc()
|
override |
Member Function Documentation
◆ adjustForOutputSpec()
|
overrideprotectedvirtual |
Change sample rate and amount of channels.
This function must be implemented when subclassing. It is used to notify subclasses about changes in source and target sample rates, as well as the amount of channels in the audio.
- Parameters
-
dstRate Target sample rate (rate being resampled to.) srcRate Source sample rate (rate being resampled from.) channels Amount of channels in both the source as well as the target audio buffers.
Implements Aulib::Resampler.
◆ doDiscardPendingSamples()
|
overrideprotectedvirtual |
Discard any internally held samples.
This function must be implemented when subclassing. It should discard any internally held samples. Note that even if you don't actually buffer any samples in your subclass but are using some external resampling library that you delegate resampling to, that external resampler might be holding samples in an internal buffer. Those will need to be discarded as well.
If none of the above applies, this can be implemented as an empty function.
Implements Aulib::Resampler.
◆ doResampling()
|
overrideprotectedvirtual |
This function must be implemented when subclassing. It must resample the audio contained in 'src' containing 'srcLen' samples, and store the resulting samples in 'dst', which has a capacity of at most 'dstLen' samples.
The 'src' buffer contains audio in either mono or stereo. Stereo is stored in interleaved format.
The source and target sample rates, as well as the amount of channels that are to be used must be those that were specified in the last call to the adjustForOutputSpec() function.
'dstLen' and 'srcLen' are both input as well as output parameters. The function must set 'dstLen' to the amount of samples that were actually stored in 'dst', and 'srcLen' to the amount of samples that were actually used from 'src'. For example, if in the following call:
dstLen = 200; srcLen = 100; doResampling(dst, src, dstLen, srcLen);
the function resamples 98 samples from 'src', resulting in 196 samples which are stored in 'dst', the function must set 'srcLen' to 98 and 'dstLen' to 196.
So when implementing this function, you do not need to worry about using up all the available samples in 'src'. Simply resample as much audio from 'src' as you can in order to fill 'dst' as much as possible, and if there's anything left at the end that cannot be resampled, simply ignore it.
Implements Aulib::Resampler.
◆ quality()
|
noexcept |
