Speech Synthesis and Processing

6.

Advanced Topics and Applications

6.1.

Speaker Recognition

6.1.1.

Speaker Identification

6.1.1.1.

Closed-set Identification

6.1.1.2.

Open-set Identification

6.1.1.3.

Text-dependent Systems

6.1.1.4.

Text-independent Systems

6.1.2.

Speaker Verification

6.1.2.1.

Authentication Applications

6.1.2.2.

Threshold Selection

6.1.2.3.

Score Normalization

6.1.2.4.

Anti-spoofing Measures

6.1.3.

Speaker Embedding Techniques

6.1.3.1.

i-vector Extraction

6.1.3.1.1.

Total Variability Space

6.1.3.1.2.

Factor Analysis

6.1.3.1.3.

Probabilistic Linear Discriminant Analysis (PLDA)

6.1.3.2.

x-vector Systems

6.1.3.2.1.

Time Delay Neural Networks (TDNNs)

6.1.3.2.2.

Statistics Pooling

6.1.3.2.3.

Deep Speaker Embeddings

6.1.4.

Channel Compensation

6.1.4.1.

Intersession Variability

6.1.4.2.

Channel Adaptation

6.1.4.3.

Domain Mismatch Handling

6.2.

Speech Enhancement

6.2.1.

Noise Reduction Techniques

6.2.1.1.

Spectral Subtraction

6.2.1.2.

Wiener Filtering

6.2.1.3.

Minimum Mean Square Error (MMSE) Estimation

6.2.1.4.

Kalman Filtering

6.2.2.

Deep Learning Enhancement

6.2.2.1.

Denoising Autoencoders

6.2.2.2.

Recurrent Neural Networks

6.2.2.3.

Generative Adversarial Networks

6.2.2.4.

Mask Estimation Networks

6.2.3.

Multi-channel Enhancement

6.2.3.1.

Beamforming Techniques

6.2.3.2.

Microphone Array Processing

6.2.3.3.

Blind Source Separation

6.2.4.

Acoustic Echo Cancellation

6.2.4.1.

Adaptive Filtering

6.2.4.2.

Double-talk Detection

6.2.4.3.

Nonlinear Echo Cancellation

6.2.5.

Dereverberation

6.2.5.1.

Room Impulse Response Modeling

6.2.5.2.

Inverse Filtering

6.2.5.3.

Statistical Reverberation Models

6.3.

Spoken Language Understanding

6.3.1.

Intent Recognition

6.3.1.1.

Classification Approaches

6.3.1.2.

Feature Engineering

6.3.1.3.

Neural Architectures

6.3.1.4.

Multi-intent Handling

6.3.2.

6.3.2.1.

Sequence Labeling

6.3.2.2.

Named Entity Recognition

6.3.2.3.

Conditional Random Fields

6.3.2.4.

Neural Sequence Models

6.3.3.

Joint Intent and Slot Modeling

6.3.3.1.

Multi-task Learning

6.3.3.2.

Attention-based Joint Models

6.3.3.3.

End-to-end SLU Systems

6.3.4.

Dialogue State Tracking

6.3.4.1.

Belief State Representation

6.3.4.2.

State Update Mechanisms

6.3.4.3.

Neural Dialogue State Trackers

6.4.

Speech Emotion Recognition

6.4.1.

Emotional Speech Databases

6.4.1.1.

Acted vs. Natural Emotions

6.4.1.2.

Annotation Schemes

6.4.1.3.

Cross-cultural Considerations

6.4.2.

Acoustic Feature Analysis

6.4.2.1.

Prosodic Features

6.4.2.2.

Spectral Features

6.4.2.3.

Voice Quality Features

6.4.2.4.

Temporal Dynamics

6.4.3.

Emotion Classification Models

6.4.3.1.

Traditional Machine Learning

6.4.3.2.

Deep Learning Approaches

6.4.3.3.

Multimodal Fusion

6.4.4.

Continuous Emotion Recognition

6.4.4.1.

Dimensional Emotion Models

6.4.4.2.

Temporal Modeling

6.4.4.3.

Real-time Processing

6.5.

Voice Conversion

6.5.1.

Parallel Voice Conversion

6.5.1.1.

Dynamic Time Warping

6.5.1.2.

Gaussian Mixture Models

6.5.1.3.

Statistical Parameter Mapping

6.5.2.

Non-parallel Voice Conversion

6.5.2.1.

CycleGAN-based Approaches

6.5.2.2.

Variational Autoencoders

6.5.2.3.

6.5.3.

Real-time Voice Conversion

6.5.3.1.

Low-latency Processing

6.5.3.2.

Streaming Algorithms

6.5.3.3.

Hardware Implementations

6.6.

Multilingual Speech Processing

6.6.1.

Cross-lingual Speech Recognition

6.6.1.1.

Multilingual Acoustic Models

6.6.1.2.

Language Identification

6.6.1.3.

Code-switching Handling

6.6.2.

Speech Translation

6.6.2.1.

Cascade Systems

6.6.2.2.

End-to-end Speech Translation

6.6.2.3.

Simultaneous Translation

6.6.3.

Multilingual TTS

6.6.3.1.

Language-specific Modeling

6.6.3.2.

Cross-lingual Voice Cloning

6.6.3.3.

Accent Modeling

Previous

5. Automatic Speech Recognition (ASR)

Go to top

Next

7. Evaluation and Quality Assessment