ELEC-E5510 - Speech Recognition D, Lecture, 26.10.2022-9.12.2022

Översikt

• Uppgifter
  In this list, the 2021 slides will be replaced by the 2022 ones after each lecture is given at latest. The titles may be identical, but the contents are improved each year based on feedback. The project works and their schedule changes each year. 

  For practicalities, e.g. regarding to the Lecture Quizzes and Exercises, check MyCourses > Course Practicalities
  

  • Välj aktivitet Lecture 1 - Feature extraction

    Lecture 1 - Feature extraction
    

  • Välj aktivitet features (2022) - final

    

    features (2022) - final Fil PDF

    • course organization
    • what is ASR
    • features of speech
    • MFCC
    • GMM
    • DNN
      
  • Välj aktivitet Lecture exercise 1: Gaussian mixture model

    

    Lecture exercise 1: Gaussian mixture model Inlämningsuppgift

    Avklarad aktivitet

    Students must

    

    Lämna in

    Please upload your answer here, e.g. as a photo, text or pdf file
    

  • Välj aktivitet Lecture 2 - Phoneme modeling

    Lecture 2 - Phoneme modeling

  • Välj aktivitet HMM (2022) part1

    

    HMM (2022) part1 Fil PDF

    • Phonemes
    • HMMs
  • Välj aktivitet HMM (2022) part2

    

    HMM (2022) part2 Fil PDF
  • Välj aktivitet Lecture 2 HMM exercise

    

    Lecture 2 HMM exercise Inlämningsuppgift

    Avklarad aktivitet

    Students must

    

    Lämna in
  • Välj aktivitet Lecture 3 - Language Modeling

    Lecture 3 - Language Modeling

  • Välj aktivitet LM (2022) part1

    

    LM (2022) part1 Fil PDF

    • lexicon
    • language modeling
    • n-grams, smoothing
  • Välj aktivitet Neural network language models (2022)

    

    Neural network language models (2022) Fil PDF

    • Intro to NNLM
    • Recurrent neural network language models
    • Long Short-Term Memory language models
    • Attention

    
    

  • Välj aktivitet Lecture 3 NNLM exercise

    

    Lecture 3 NNLM exercise Inlämningsuppgift

    Avklarad aktivitet

    Students must

    

    Lämna in
  • Välj aktivitet Lecture 4 - Continuous speech and decoding

    Lecture 4 - Continuous speech and decoding
    

  • Välj aktivitet LVCSR (2022) - Final

    

    LVCSR (2022) - Final Fil PDF

    • recognition in continuous speech
    • token passing decoder
    • improving the recognition performance and speed
      
    • measuring the recognition performance
      
  • Välj aktivitet Lecture exercise 4: Token passing decoder

    Show more Show less

    Tillgänglig om någon av: You are a(n) Lärare ...

    Tillgänglig om någon av:

    • You are a(n) Lärare
    • You are a(n) Teacher (MC)
    • You are a(n) Studerande

    

    Lecture exercise 4: Token passing decoder Inlämningsuppgift

    Fill in the last column with final probabilities of the tokens, select the best token and output the corresponding state sequence!
    

    The goal is to verify that you have the learned the idea of the Token passing decoder. The extremely simplified HMM system is almost the same as in the 2B Viterbi algorithm exercise. The observed "sounds" are just quantified to either "A" or "B" with given probabilities in states S0 and S1.  Now the task is to find the most likely state sequence that can produce the sequence of sounds A, A, B using a simple language model (LM). The toy LM used here is a look-up table that tells probabilities for different state sequences, (0,1), (0,0,1) etc., up to 3-grams.
    

    Hint: You can either upload an edited source document, a pdf file, a photo of your notes or a text file with numbers. Whatever is easiest for you. To get the activity point the answer does not have to be correct.

    
    

  • Välj aktivitet Lecture 5 - End-to-end ASR with deep neural networ...

    Lecture 5 - End-to-end ASR with deep neural networks
    

    
    
  • Välj aktivitet Lecture 5 Slides

    

    Lecture 5 Slides Fil PDF
  • Välj aktivitet Lecture 5 exercise

    

    Lecture 5 exercise Inlämningsuppgift

    Avklarad aktivitet

    Students must

    

    Lämna in
  • Välj aktivitet Lecture 5 slides (2022): About group works (Final)

    

    Lecture 5 slides (2022): About group works (Final) Fil PDF

    Preparation for the seminar sessions during the last two weeks

    Updated presentation schedule for the groups
    

  • Välj aktivitet Program for the last two weeks

    Program for the last two weeks
    

  • Välj aktivitet Wed 30 November: Presentations 1

    Wed 30 November: Presentations 1
    

  • Välj aktivitet Fri 2 December: Presentations 2 - 5

    Fri 2 December: Presentations 2 - 5
    

  • Välj aktivitet Wed 7 December: Presentations 6 - 9

    Wed 7 December: Presentations 6 - 9
    

  • Välj aktivitet Fri 9 December: Presentations 10 - 13

    Fri 9 December: Presentations 10 - 13
    

  • Välj aktivitet Text and media area

    
    

  • Välj aktivitet PresentersTwo days before (or earlier if possible)...

    Presenters
    

    • Two days before (or earlier if possible): Select one article for others to read and send the link to everybody in MyCourses discussion forum
    • One day before: Upload your slides in MyCourses. The latest version of slides will be published for others in MyCourses. You can also share a draft of the slides or a link in the discussion forum.
    • Practise to make sure that you will not exceed the 20 mins limit
    • Remember your “audience” duties for the other talks of your day

    Audience
    For each talk do this:
    

    • One day before: Read the provided articles and prepare one question to ask for each talk
    • Follow the talk and the slides and ask your question in chat
    • After the talk (max 1 day): Submit feedback (one for each talk) in MyCourses (all fields are required) to get activity points. The anonymous feedback (pros and cons) will be later shown to the presenters