PyAnnoteOptions
Options & Configuration · AiDotNet.Audio.Speaker
Configuration options for the pyannote 3.x speaker diarization model.
Inheritance: ModelOptions → PyAnnoteOptions
For Beginners
pyannote is a system that figures out “who spoke when” in a recording with multiple speakers. It splits audio into segments, assigns each to a speaker, and can even detect when two people talk at the same time. It’s widely used for meeting transcription, podcast processing, and call analytics.
How It Works
pyannote.audio 3.x (Plaquet & Bredin, ASRU 2023) is a state-of-the-art speaker diarization pipeline using end-to-end neural segmentation with PyanNet architecture. It segments audio into speaker turns and supports overlapping speech detection. Achieves 11.2% DER on AMI Mix-Headset benchmark.
Constructors
| Constructor | Summary |
|---|---|
PyAnnoteOptions() |
Properties
| Property | Summary |
|---|---|
OnnxModelOptions OnnxOptions { get; set; } |
Gets or sets the ONNX runtime options. |
bool EnableOverlapDetection { get; set; } |
Gets or sets whether overlapping speech detection is enabled. |
double ChunkDurationSeconds { get; set; } |
Gets or sets the segmentation chunk duration in seconds. |
double ChunkStepSeconds { get; set; } |
Gets or sets the step between consecutive chunks in seconds. |
double ClusteringThreshold { get; set; } |
Gets or sets the clustering threshold for speaker assignment. |
double DropoutRate { get; set; } |
Gets or sets the dropout rate. |
double LearningRate { get; set; } |
Gets or sets the learning rate for training. |
double MinSegmentDuration { get; set; } |
Gets or sets the minimum segment duration in seconds. |
double WeightDecay { get; set; } |
Gets or sets the weight decay for regularization. |
int EmbeddingDim { get; set; } |
Gets or sets the embedding dimension for speaker embeddings. |
int FftSize { get; set; } |
Gets or sets the FFT window size in samples. |
int HopLength { get; set; } |
Gets or sets the hop length between frames in samples. |
int LSTMHiddenSize { get; set; } |
Gets or sets the LSTM hidden size for the segmentation model. |
int LinearDim { get; set; } |
Gets or sets the linear layer hidden dimension after LSTM. |
int MaxSpeakersPerChunk { get; set; } |
Gets or sets the maximum number of speakers per chunk. |
int NumLSTMLayers { get; set; } |
Gets or sets the number of LSTM layers. |
int NumMels { get; set; } |
Gets or sets the number of mel filterbank channels. |
int SampleRate { get; set; } |
Gets or sets the expected audio sample rate in Hz. |
int SincNetFilters { get; set; } |
Gets or sets the number of SincNet filters in the first layer. |
int? MaxSpeakers { get; set; } |
Gets or sets the maximum number of speakers (null for auto-detection). |
string EmbeddingModelPath { get; set; } |
Gets or sets the path to a separate embedding model. |
string ModelPath { get; set; } |
Gets or sets the path to the ONNX model file. |