speechbrain.lobes.models¶
Package defining neural netword models (CRDNN, Xvectors …)
A combination of Convolutional, Recurrent, and Fully-connected networks. |
|
The SpeechBrain implementation of ContextNet by https://arxiv.org/pdf/2005.03191.pdf |
|
A popular speaker recognition and diarization model. |
|
This lobes replicate the encoder first introduced in ESPNET v1 |
|
Generator and discriminator used in MetricGAN |
|
Implementation of a Recurrent Language Model. |
|
Vanilla Neural Network for simple tests. |
|
A popular speaker recognition and diarization model. |
|
Implementation of a popular speech separation model. |
|
This is a module to ensemble a convolution (depthwise) encoder with or without residule connection. |
|
Library to support dual-path speech separation. |
|
This lobe enables the integration of fairseq pretrained wav2vec models. |
|
This lobe enables the integration of huggingface pretrained wav2vec2 models. |
High level processing blocks. |