Nay San's repositories
lnl-examples
PyTorch Lightning and Lhotse examples
PTL2-DS2ish
A toy repository for using PyTorch Lightning 2.x to train an adapted DeepSpeech 2 model
scriptable_hubert_encoder
Development repository for experimenting with a scriptable (and crammable) HuBERT encoder
w2v2-10min-exps
Experiments with wav2vec 2.0 models involving only 10 minutes of transcribed speech
w2v2-batch-size
Code for paper "The effect of batch size on contrastive self-supervised speech representation learning"
asr-dataset-prep
Scripts for preparing datasets for automatic speech recognition
E2E-language-diarization-transfer
Source code of paper <End-to-End Language Diarization for Bilingual Code-switching Speech>
glottospace
Geospatial analysis of linguistic data
jupyter-book
Create beautiful, publication-quality books and documents from computational content.
lightning-speech-sampling
Try out different samplers for speech data with PyTorch Lightning
OPUS-MT-train
Training open neural machine translation models
slasr-scripts
Data processing scripts for SLASR project
w2v2-10min-replication
Replicate training wav2vec 2.0 model on just 10 minutes of Librispeech data
w2v2-hf-pretrain-test
Testing wav2vec 2.0 pre-training with HuggingFace
Wav2vec2.0
Implementation of the paper "wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations" in Pytorch.