AcousticModelOptions

Options & Configuration · AiDotNet.TextToSpeech.Classic

Base configuration options for classic acoustic TTS models.

Inheritance: ModelOptions → TtsModelOptions → AcousticModelOptions

Known derived types: AlignTTSOptions, DeepVoice3Options, FastSpeech2Options, FastSpeechOptions, GlowTTSOptions, GradTTSOptions, NonAttentiveTacotronOptions, PortaSpeechOptions, ProDiffOptions, SpeedySpeechOptions, Tacotron2Options, TacotronOptions, TransformerTTSOptions

For Beginners

These options configure the AcousticModel model. Default values follow the original paper settings.

Constructors

Constructor Summary
AcousticModelOptions() Initializes a new instance with default values.
AcousticModelOptions(AcousticModelOptions other) Initializes a new instance by copying from another instance.

Properties

Property Summary
bool UsePostnet { get; set; } Gets or sets whether to use a postnet for mel refinement.
int DecoderDim { get; set; } Gets or sets the decoder output dimension (mel channels).
int EncoderDim { get; set; } Gets or sets the encoder embedding dimension.
int OutputsPerStep { get; set; } Gets or sets the number of mel frames generated per decoder step (reduction factor).
int PostnetDim { get; set; } Gets or sets the postnet embedding dimension.
int PostnetLayers { get; set; } Gets or sets the number of postnet convolution layers.
string[] TextSymbols { get; set; } Ordered symbol inventory used to train the text embedding.