renee_thesis

All the code for my thesis on improving speech recognition systems for children. I am currently working on using Facebook AI's wav2vec 2.0.

Steps for kaldi: Make sure kaldi is installed first before proceeding.

1. Download all the required datasets
2. In s5/run.sh modify [DATASET NAME]_ROOT to point to the main directory of the dataset
3. ./run.sh This will build the HMM-GMM model
4. After completed successfully run local/nnet3/run_tdnn_delta.sh for the TDNN model THe script s5/clean.sh will remove file created from s5/run.sh so that you can train the models again.

Steps for wav2vec 2.0: [WIP] Make sure PyTorch and fairseq are installed first before proceeding.

1. In s5/wav2vec_projects run the various run_* scripts.

Install Kaldi: Refer to: https://kaldi-asr.org/doc/tutorial_setup.html

1. git clone https://github.com/kaldi-asr/kaldi.git
2. Look at the kaldi/INSTALL file and follow the instructions there
3. Download SRILM by running kaldi/tools/install_srilm.sh

Install PyTorch & fairseq PyTorch: https://pytorch.org/get-started/locally/ fairseq: https://www.folio3.ai/blog/fairseq/

Initialising katana:

To execute the steps using the supercomputer katana.

1. In Windows PuTTY: use host name = katana1.restech.unsw.edu.au and log in using zID and password In Linux: ssh zID@katana1.restech.unsw.edu.au in terminal. Or, use the alias katana.
2. Create a new screen using screen -S nameOfSession (the screen I'm using is called 'thesis')
3. Request an interactive GPU node using qsub -I -l select=1:ngpus=1:ncpus=8:mem=46gb,walltime=12:00:00. Once the node is ready, you are now in the node. The terminal will show (zID@kxxx), where kxxx is your node.
4. Now you are inside the screen, and inside the GPU node. Run whatever process you need.
5. To load modules and go inside the thesis virtual environment, run startup.sh

Note: If there is an error message saying Permission Denied when running a script, use chmod u+x -R /path/to/directory to change the permissions of all the files in the directory so that you have permission to execute.

To install any python packages not in katana, use a virtual environment: https://packaging.python.org/guides/installing-using-pip-and-virtual-environments/ The virtual environment I am using currently is kaldi/egs/renee_thesis/thesis_env source thesis_env/bin/activate To list all the packages in this virtual environment use pip list

Connecting to GitHub

• To start a new git repository follow the instructions here: https://kbroman.org/github_tutorial/pages/init.html

• When ready to push changes:

  git add -A Adds all files

  git commit -m "Commit message here" Commits changes with a message

  git push origin master Pushes changes

• If there are Permission Denied errors follow the instructions here: https://gist.github.com/adamjohnson/5682757

• To check the status use git status

Leaving katana Asumming you are inside a screen, and inside a requested GPU node.

1. CtrlA D to detach from the screen session.
2. exit to logout of the katana session.

Returning to katana

1. ssh zID@katana1.restech.unsw.edu.au in terminal. Or, use the alias katana.
2. Go back to your screen screen -r nameOfSession eg. screen -r ogi

Useful katana screen things

• To create a new window (tab) within a screen, use CtrlA C
• To go to next and previous windows, use CtrlA N and CtrlA P respectively.
• To check if you are in a screen, type the command echo $TERM
• To list your screen, type command screen -list
• To detach a screen remotely, find the screen name using screen -list and then screen -d [name of screen]

Bash things

• Create aliases in ~./bashrc by editing ~./bashrc and then running source ~./bashrc: https://linuxize.com/post/how-to-create-bash-aliases/
• Current aliases: t2renee_thesis t2chacmod
• Screen tabs: https://unix.stackexchange.com/questions/26248/tabs-when-using-screen/319364