BandSplitRNNEnhancerOptions
Options & Configuration · AiDotNet.Audio.Enhancement
Configuration options for the Band-Split RNN enhancement model.
Inheritance: ModelOptions → BandSplitRNNEnhancerOptions
For Beginners
Imagine you’re in a noisy room trying to hear someone speak. Band-Split RNN works like having multiple specialized listeners, each focused on a different pitch range (bass, midrange, treble). Each listener cleans up their range, and then they combine their results. This divide-and-conquer approach works very well because different types of noise affect different frequency ranges.
How It Works
Band-Split RNN (Luo & Yu, 2023) splits the spectrogram into non-overlapping frequency bands, processes each band with a shared RNN, then fuses the bands. Originally designed for music source separation, it also excels at speech enhancement by treating noise as a “source” to separate. Band-Split RNN achieves state-of-the-art on both music separation and speech enhancement benchmarks.
Constructors
| Constructor | Summary |
|---|---|
BandSplitRNNEnhancerOptions() |
Initializes options with the paper-default Band-Split RNN configuration. |
BandSplitRNNEnhancerOptions(BandSplitRNNEnhancerOptions other) |
Initializes a new instance by deep-copying all enhancer settings. |
Properties
| Property | Summary |
|---|---|
OnnxModelOptions OnnxOptions { get; set; } |
Gets or sets the ONNX runtime options. |
double DropoutRate { get; set; } |
Gets or sets the dropout rate. |
double LearningRate { get; set; } |
Gets or sets the learning rate. |
int BandRnnHiddenSize { get; set; } |
Gets or sets the RNN hidden size per band. |
int FFTSize { get; set; } |
Gets or sets the FFT size. |
int FusionDim { get; set; } |
Gets or sets the band fusion hidden dimension. |
int HopLength { get; set; } |
Gets or sets the hop length. |
int NumBands { get; set; } |
Gets or sets the number of frequency bands. |
int NumFreqBins { get; set; } |
Gets or sets the number of frequency bins. |
int NumRnnLayers { get; set; } |
Gets or sets the number of RNN layers per band. |
int SampleRate { get; set; } |
Gets or sets the expected audio sample rate in Hz. |
string ModelPath { get; set; } |
Gets or sets the path to the ONNX model file. |
string Variant { get; set; } |
Gets or sets the model variant (“small”, “medium”, or “large”). |