cvqluu
Research Scientist working on ML/AI for speech tech. PhD from the University of Edinburgh, CSTR
DeepgramEdinburgh, United Kingdom
Pinned Repositories
Angular-Penalty-Softmax-Losses-Pytorch
Angular penalty loss functions in Pytorch (ArcFace, SphereFace, Additive Margin, CosFace)
dropclass_speaker
DropClass and DropAdapt - repository for the paper accepted to Speaker Odyssey 2020
Factorized-TDNN
PyTorch implementation of the Factorized TDNN (TDNN-F) from "Semi-Orthogonal Low-Rank Matrix Factorization for Deep Neural Networks" and Kaldi
GE2E-Loss
Pytorch implementation of Generalized End-to-End Loss for speaker verification
MTL-Speaker-Embeddings
Code for the paper: "Leveraging speaker attribute information using multi task learning for speaker verification and diarization" presented at Interspeech 2021
nn-similarity-diarization
Neural network based similarity scoring for diarization (pytorch implementation of "LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization")
simple_diarizer
Simplified diarization pipeline using some pretrained models - audio file to diarized segments in a few lines of code
splitdim_disentangle
Disentangling speaker embeddings using multi-task learning and adversarial training
TDNN
Time delay neural network (TDNN) implementation in Pytorch using unfold method
animalsupportbot
Reddit bot that replies with the most relevant counterargument to debunk common anti-vegan myths. It replies to a parent comment when summoned via mentions. Powered by Artificial Intelligence and open-source.
cvqluu's Repositories
cvqluu/Angular-Penalty-Softmax-Losses-Pytorch
Angular penalty loss functions in Pytorch (ArcFace, SphereFace, Additive Margin, CosFace)
cvqluu/TDNN
Time delay neural network (TDNN) implementation in Pytorch using unfold method
cvqluu/Factorized-TDNN
PyTorch implementation of the Factorized TDNN (TDNN-F) from "Semi-Orthogonal Low-Rank Matrix Factorization for Deep Neural Networks" and Kaldi
cvqluu/simple_diarizer
Simplified diarization pipeline using some pretrained models - audio file to diarized segments in a few lines of code
cvqluu/GE2E-Loss
Pytorch implementation of Generalized End-to-End Loss for speaker verification
cvqluu/nn-similarity-diarization
Neural network based similarity scoring for diarization (pytorch implementation of "LSTM based Similarity Measurement with Spectral Clustering for Speaker Diarization")
cvqluu/MTL-Speaker-Embeddings
Code for the paper: "Leveraging speaker attribute information using multi task learning for speaker verification and diarization" presented at Interspeech 2021
cvqluu/dropclass_speaker
DropClass and DropAdapt - repository for the paper accepted to Speaker Odyssey 2020
cvqluu/splitdim_disentangle
Disentangling speaker embeddings using multi-task learning and adversarial training
cvqluu/gitignore
A collection of useful .gitignore templates
cvqluu/cvqluu.github.io
cvqluu/face.evoLVe.PyTorch
🔥🔥High-Performance Face Recognition Library on PyTorch🔥🔥
cvqluu/mlpractical-1
Machine Learning Practical course repository
cvqluu/mpv
🎥 Command line video player
cvqluu/supreme_court_transcripts
cvqluu/zmk-config
cvqluu/zmk-config-ckp
zmk-config for the polarity works ckp keyboard seties