Robert Flynn's repositories
hydra-linear-attention
Implementation of: Hydra Attention: Efficient Attention with Many Heads (https://arxiv.org/abs/2209.07484)
long-context-asr
Code for the paper: How Much Context Does My Attention-Based ASR System Need?
causal-transformer
A causal transformer based LM.
CNN-LCP-Shared-Task-2021
Code for Lexical Complexity Prediction Shared Task
Self-Train-Before-You-Transcribe
Code for Interspeech 2024 paper: Self-Train Before You Transcribe
batchrenorm
Batch Renormalization in Pytorch - added masking
ionic-app-base
A base starting point for Ionic, with Cordova, Bower, and Gulp.
lhotse
Tools for handling speech data in machine learning projects.
MakeProgress
This is a template, which lays out a structure for a main project. It can be used by any team working on an MMU LiveProject.
NeMo
NeMo: a toolkit for conversational AI
synthetic-tasks
Collection of synthetic tasks (wip) for testing neural sequence models
flip-with-mask-pytorch
example of flip operation that keeps masked indices at the end of the sequence
number_conversion
For use with num2words. Convert a document with numbers in to a document with just words
pyctcdecode
A fast and lightweight python-based CTC beam search decoder for speech recognition.
robflynnyh
Config files for my GitHub profile.
Sheffield-Model-Inference
An inference pipeline to simplify usage of large publicly available pre-trained ML model on HPC
vector-quantize-pytorch
Vector Quantization, in Pytorch