TDTDecoderOptions

Options & Configuration · AiDotNet.SpeechRecognition.Streaming

Options for TDT.

Inheritance: ModelOptions → TDTDecoderOptions

For Beginners

These options set the size of the Conformer encoder and decoder, the feature front end and the training augmentation.

How It Works

The defaults: 17 Conformer blocks, width 512, 8 heads, a 31-wide depthwise kernel, a 640-unit single-layer LSTM prediction network and a 640-wide joint network.

Constructors

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

Properties

Property Summary
OnnxModelOptions OnnxOptions { get; set; } ONNX runtime options.
double RopeTheta { get; set; } Rotary base of the attention’s relative position encoding, standing in for the paper’s relative sinusoidal encoding (section 2.1).
double TimeMaskRatio { get; set; } Largest time mask as a fraction of the utterance (section 3.1: pS = 0.05).
int ConvKernel { get; set; } Depthwise convolution kernel (Conformer-Large: 31).
int EncoderDim { get; set; } Conformer width.
int FeedForwardMultiplier { get; set; } Expansion of each half-step feed-forward module (section 2.3: 4).
int FrequencyMaskWidth { get; set; } Largest frequency mask, in mel bins (section 3.1: F = 27).
int FrequencyMasks { get; set; } SpecAugment frequency masks (section 3.1: 2).
int JointDim { get; set; } Width of the joint network.
int MaxAudioLengthSeconds { get; set; } Longest audio a transcription covers, in seconds; longer input is truncated.
int NumAttentionHeads { get; set; } Attention heads per block.
int NumEncoderLayers { get; set; } Conformer blocks.
int NumMels { get; set; } Log-mel bins (section 3.1: 80-channel filterbanks over 25 ms windows with a 10 ms stride).
int PredictionDim { get; set; } LSTM units of the prediction network (section 3.1: a single-layer LSTM of 640 units for Conformer M and L, 320 for S).
int PredictionLayers { get; set; } LSTM layers of the prediction network (section 3.1: 1).
int SampleRate { get; set; } Audio sample rate in Hz.
int TimeMasks { get; set; } SpecAugment time masks (section 3.1: 10).
int VocabSize { get; set; } Output vocabulary size, blank (token 0) included (1024 SentencePiece tokens).
int[] Durations { get; set; } The frame durations the joint network chooses among (Xu et al.
string Language { get; set; } Language reported for transcriptions.
string ModelPath { get; set; } Path to an exported ONNX model, when one is used.
string[] Vocabulary { get; set; } The text of each token; tokens past the list decode as Unicode code points, and tokens in angle brackets decode to nothing.