MayaFlux 1.0.0
Digital-First Multimedia Processing Framework
Loading...
Searching...
No Matches

◆ frame_audio_stage()

DependencyStage MayaFlux::Yantra::TensorKernels::frame_audio_stage ( uint64_t samples,
uint64_t frames,
uint32_t chunks,
uint32_t chunk_size,
uint32_t context,
uint32_t pad )
inlinenodiscard

One nn_frame_audio dispatch: cut a sample stream into overlapping frames, reflect padded on the right.

samples holds context + chunks * chunk_size values. Frame n is the context + chunk_size values starting at n * chunk_size, followed by pad values that mirror the frame about its last value without repeating it. pad must be smaller than context + chunk_size.

Parameters
samplesDevice address of the samples.
framesDevice address of the frames, [chunks, frame_audio_length].
Exceptions
std::invalid_argumentWhen chunks or chunk_size is 0 or pad is a frame or more.

Definition at line 267 of file Stages.hpp.

269{
270 if (chunks == 0 || chunk_size == 0 || pad >= context + chunk_size) {
271 detail::refuse("Frame audio needs chunks, a chunk size and a pad smaller than one frame");
272 }
273 const FrameAudioPC pc { .samples = samples, .frames = frames, .chunks = chunks, .chunk_size = chunk_size, .context = context, .pad = pad };
275 detail::groups_for(static_cast<size_t>(chunks) * frame_audio_length(chunk_size, context, pad), element_workgroup));
276}
DependencyStage make_stage(const char *shader, uint32_t local, const PC &pc, uint32_t groups)
Definition Stages.hpp:48
void refuse(std::string_view message)
Definition Layers.hpp:74
uint32_t groups_for(size_t invocations, uint32_t local)
Definition Stages.hpp:21
constexpr uint32_t frame_audio_length(uint32_t chunk_size, uint32_t context, uint32_t pad)
Length of one frame nn_frame_audio writes.
Definition Layers.hpp:120
constexpr const char * frame_audio_shader
Definition Layouts.hpp:23
constexpr uint32_t element_workgroup
Invocations per workgroup of the element-wise kernels.
Definition Layouts.hpp:15

References element_workgroup, frame_audio_length(), frame_audio_shader, MayaFlux::Yantra::TensorKernels::detail::groups_for(), MayaFlux::Yantra::TensorKernels::detail::make_stage(), and MayaFlux::Yantra::TensorKernels::detail::refuse().

Referenced by MayaFlux::Yantra::SileroVadModel::Impl::run().

Here is the call graph for this function:
Here is the caller graph for this function: