AudioNeuralNetworkOptions
Options & Configuration · AiDotNet.Models.Options
Base configuration options for audio neural network models.
Inheritance: ModelOptions → NeuralNetworkOptions → ModelHyperparameterOptions → AudioNeuralNetworkOptions
Known derived types: AudioGenOptions, CLAPModelOptions, ConvTasNetOptions, DCCRNOptions, DeepFilterNetOptions, NeuralNoiseReducerOptions, PANNsModelOptions, SileroVadOptions, Wav2Vec2ModelOptions, WhisperOptions
How It Works
Every options class under src/Audio already derives from this, so it is where the
audio models’ validation helpers belong — the same relationship
DocumentNeuralNetworkOptions has with the document models.
Unlike the other families, this base declares NO signal properties. The audio options
classes already declared their own SampleRate, NumMels, FftSize,
HopLength and the rest before any shared base existed — around thirty of them — so
declaring those names here would hide each leaf’s property rather than share one.
Constructors
| Constructor | Summary |
|---|---|
AudioNeuralNetworkOptions() |
Properties
| Property | Summary |
|---|---|
bool ReturnTimestamps { get; set; } |
Gets or sets whether transcription returns per-segment timestamps by default. |