Speech Synthesis and Processing

  1. Automatic Speech Recognition (ASR)
    1. ASR System Architecture
      1. Feature Extraction Pipeline
        1. Preprocessing Steps
          1. Feature Computation
            1. Feature Normalization
              1. Feature Transformation
              2. Acoustic Modeling
                1. Model Architecture Design
                  1. Training Procedures
                    1. Model Adaptation Techniques
                    2. Language Modeling
                      1. Vocabulary Design
                        1. Text Preprocessing
                          1. Model Training and Evaluation
                          2. Decoding Process
                            1. Search Space Definition
                              1. Pruning Strategies
                                1. Hypothesis Scoring
                              2. Traditional ASR Approaches
                                1. Hidden Markov Models (HMMs)
                                  1. HMM Topology Design
                                    1. State Emission Modeling
                                      1. Transition Probability Modeling
                                        1. Training Algorithms
                                          1. Baum-Welch Algorithm
                                            1. Viterbi Training
                                          2. Gaussian Mixture Models (GMMs)
                                            1. GMM Parameter Estimation
                                              1. Component Selection
                                                1. Adaptation Techniques
                                                2. GMM-HMM Systems
                                                  1. System Architecture
                                                    1. Training Pipeline
                                                      1. Decoding Process
                                                    2. Deep Learning ASR
                                                      1. Deep Neural Network Acoustic Models
                                                        1. Feedforward Networks
                                                          1. Convolutional Neural Networks
                                                            1. Recurrent Neural Networks
                                                              1. Hybrid DNN-HMM Systems
                                                              2. End-to-End ASR Models
                                                                1. Connectionist Temporal Classification (CTC)
                                                                  1. CTC Loss Function
                                                                    1. Alignment-free Training
                                                                      1. CTC Decoding Algorithms
                                                                      2. Attention-based Models
                                                                        1. Encoder-Decoder Architecture
                                                                          1. Attention Mechanisms
                                                                            1. Listen, Attend and Spell (LAS)
                                                                            2. RNN Transducer (RNN-T)
                                                                              1. Transducer Architecture
                                                                                1. Streaming Recognition
                                                                                  1. Training and Inference
                                                                                2. Transformer-based ASR
                                                                                  1. Self-attention Mechanisms
                                                                                    1. Conformer Architecture
                                                                                      1. Wav2Vec 2.0
                                                                                        1. Whisper Architecture
                                                                                      2. Language Modeling
                                                                                        1. N-gram Language Models
                                                                                          1. Maximum Likelihood Estimation
                                                                                            1. Smoothing Techniques
                                                                                              1. Add-one Smoothing
                                                                                                1. Good-Turing Smoothing
                                                                                                  1. Kneser-Ney Smoothing
                                                                                                  2. Backoff and Interpolation
                                                                                                  3. Neural Language Models
                                                                                                    1. Feedforward Neural LMs
                                                                                                      1. Recurrent Neural LMs
                                                                                                        1. LSTM Language Models
                                                                                                          1. Transformer Language Models
                                                                                                          2. Domain Adaptation
                                                                                                            1. Model Interpolation
                                                                                                              1. Model Adaptation Techniques
                                                                                                                1. Transfer Learning

                                                                                                            Previous

                                                                                                            4. Speech Synthesis (Text-to-Speech)

                                                                                                            Go to top

                                                                                                            Next

                                                                                                            6. Advanced Topics and Applications