Jiyang (Mark) Tang's repositories
std-mandarin-kaldi
Script for training a non-chain tdnn model for standard mandarin GOP scoring. The training data is from a filtered subset of AISHELL2 and MAGICDATA which contain (relatively) standard Mandarin pronunciation
speech-recognition
GMM-HMM Continuous ASR Using Python and Numpy
aidatatang_force_align
Perform force alignment on Mandarin data using aidatatang pretrained model at https://kaldi-asr.org/models/m10
capt-public
Public version of my Computer-Aided Pronunciation Training (CAPT) system (server)
tjy_vic3_fix
TJY's Victoria 3 Improvements Mod
ali_to_phone
Extract phone-level alignment and phonemic transcript from kaldi ali.*.gz files
blender-projects
Muscle memory of blender shortcuts never fades away
dance-classifier
CS302 CV Final Project
float_repr
Utilities for viewing floating point representation
gitignore
The largest collection of useful .gitignore templates
kaldi
Kaldi fork used for pronunciation evaluation experiments
kaldi-align-to-phones
Use kaldi pretrained nnet3 model to align individual sentences and get phone-level transcripts
parallel_cpp
Some fun experiments with threads and processes
STATS403_acoustic_unit_discovery
Reproducing the acoustic unit discovery experiments (https://github.com/bshall/ZeroSpeech), testing on zerospeech2020 (https://github.com/bootphon/zerospeech2020), and using it to do ASR just for fun
zerospeech2020
Modified version of "Python package for the Zero Speech Challenge 2020". Some errors are fixed for zerospeech2020 evaluation task, and dependencies are included as submodules