SpeakerLMOptions

Options & Configuration · AiDotNet.Audio.Speaker

Configuration options for the SpeakerLM model.

Inheritance: ModelOptions → SpeakerLMOptions

For Beginners

SpeakerLM treats speaker recognition like a language problem - it “reads” speaker characteristics the way a language model reads words. This lets it handle complex multi-speaker scenarios where multiple people are talking.

How It Works

SpeakerLM (2024) uses a language model backbone for speaker understanding tasks including speaker verification, diarization, and speaker-attributed transcription. It processes speaker embeddings as tokens in a sequence, enabling multi-speaker understanding in a unified framework.

Constructors

Constructor Summary
SpeakerLMOptions()

Properties

Property Summary
OnnxModelOptions OnnxOptions { get; set; } Gets or sets the ONNX runtime options.
double DefaultThreshold { get; set; } Gets or sets the default verification threshold (cosine similarity).
double DropoutRate { get; set; } Gets or sets the dropout rate.
double LearningRate { get; set; } Gets or sets the learning rate.
double MinDurationSeconds { get; set; } Gets or sets the minimum audio duration in seconds for reliable embedding extraction.
int EmbeddingDim { get; set; } Gets or sets the speaker embedding dimension.
int LMHiddenDim { get; set; } Gets or sets the language model hidden dimension.
int MaxSpeakers { get; set; } Gets or sets the maximum number of speakers.
int NumHeads { get; set; } Gets or sets the number of attention heads.
int NumLMLayers { get; set; } Gets or sets the number of LM layers.
int NumMels { get; set; } Gets or sets the number of mel-frequency bins for audio preprocessing.
int SampleRate { get; set; } Gets or sets the 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.