sovse / dla

Deep learning for audio processing

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

logo5v1

Deep Learning for Audio (DLA)

  • Lecture and seminar materials for each week are in ./week* folders, see README.md for materials and instructions
  • Any technical issues, ideas, bugs in course materials, contribution ideas - add an issue
  • The current version of the course is conducted in autumn 2021 at the CS Faculty of HSE

Syllabus

  • week01 Introduction to Digital Signal Processing

    • Lecture: Introduction
    • Seminar: Intro in pytorch
  • week02 Introduction to Digital Signal Processing

    • Lecture: Signals, Fourier transform, Spectrograms, MelScale, MFCC and etc
    • Seminar: torchaudio, spoken digit classification
  • week03 Automatic Speech Recognition I

    • Lecture: Metrics, Attention, CTC, LAS, BeamSearch
    • Seminar: Audio augmentations, CTC decoding, CTC BeamSearch
  • week04 Automatic Speech Recognition II

    • Lecture: RNN-T, LM-fusion, BPE
    • Seminar: W&B tutorial, homework barebones overview
  • week05 Speaker verification and identification

    • Lecture: Metric Learning: Cosine, Contrastive, Triplet Losses. Angular Softmax. ArcFace
    • Seminar: ---
  • week06 Key-word spottind (KWS)

    • Lecture: (DNN, CNN, RNN+Attention) based KWS, SVDF, Orthogonality Regularization and other Tricks
    • Seminar: Speeding Up NNs: Tensor Decomposition, Quantization, Pruning, Distilation and Architecture Design

Homeworks

  • ASR Implementation of ASR model

  • KWS Implementation of KWS model

Contributors & course staff

Course materials and teaching performed by

About

Deep learning for audio processing

License:MIT License


Languages

Language:Jupyter Notebook 100.0%