cvqluu/TDNN

Stars
189
Rank 204,649 (Top 5 %)
Language
Python
Created over 5 years ago
Updated about 5 years ago

cvqluu/TDNN

cvqluu

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Time delay neural network (TDNN) implementation in Pytorch using unfold method

TDNN

Simple Time Delay Neural Network (TDNN) implementation in Pytorch. Uses the unfold method to slide over an input sequence.

[1] https://www.danielpovey.com/files/2015_interspeech_multisplice.pdf

Factorized TDNN (TDNN-F)

I've also implemented the Factorized TDNN from Kaldi (TDNN-F) in PyTorch here: https://github.com/cvqluu/Factorized-TDNN

Usage

To recreate the TDNN part of the x-vector network in [2]:

from tdnn import TDNN

# Assuming 24 dim MFCCs per frame

frame1 = TDNN(input_dim=24, output_dim=512, context_size=5, dilation=1)
frame2 = TDNN(input_dim=512, output_dim=512, context_size=3, dilation=2)
frame3 = TDNN(input_dim=512, output_dim=512, context_size=3, dilation=3)
frame4 = TDNN(input_dim=512, output_dim=512, context_size=1, dilation=1)
frame5 = TDNN(input_dim=512, output_dim=1500, context_size=1, dilation=1)

# Input to frame1 is of shape (batch_size, T, 24)
# Output of frame5 will be (batch_size, T-14, 1500)

[2] https://www.danielpovey.com/files/2018_icassp_xvectors.pdf

Angular-Penalty-Softmax-Losses-Pytorch

Angular penalty loss functions in Pytorch (ArcFace, SphereFace, Additive Margin, CosFace)

Factorized-TDNN

PyTorch implementation of the Factorized TDNN (TDNN-F) from "Semi-Orthogonal Low-Rank Matrix Factorization for Deep Neural Networks" and Kaldi

simple_diarizer

Simplified diarization pipeline using some pretrained models - audio file to diarized segments in a few lines of code

GE2E-Loss

Pytorch implementation of Generalized End-to-End Loss for speaker verification

nn-similarity-diarization

Neural network based similarity scoring for diarization (pytorch implementation of "LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization")

MTL-Speaker-Embeddings

Code for the paper: "Leveraging speaker attribute information using multi task learning for speaker verification and diarization" presented at Interspeech 2021

dropclass_speaker

DropClass and DropAdapt - repository for the paper accepted to Speaker Odyssey 2020

splitdim_disentangle

Disentangling speaker embeddings using multi-task learning and adversarial training