|
MayaFlux 1.0.0
Digital-First Multimedia Processing Framework
|
|
nodiscard |
Where the recurrent output of the latest run is on the device.
The buffer holds hidden_width float32 values per chunk, chunk after chunk: what the network knows about the stream at each chunk, from which its last layer derives the probability. It was trained only to tell speech from other sound, so nothing else is claimed of it. It is overwritten by the next run and follows the model when it grows.
Definition at line 251 of file SileroVadModel.cpp.
References m_impl.
Referenced by MayaFlux::Yantra::VadAnalyzer< std::vector< Kakshya::DataVariant >, Eigen::MatrixXd >::hidden_buffer().