• Awesome Speaker In Speech Field
  • Book recommendations
  • Speaker Recognition/Verification:
    • Toolkit
    • Speaker Models
    • Aggregation Layers
    • Datasets
  • Voice Vonversion
  • Target Speaker Extraction
  • Speaker Diarization
  • Spoofing

Awesome Speaker In Speech Field

Hi, everyone! I’m Junjie Li [Homepage], currently a Ph.D. student at Hong Kong Polytechnic University (PolyU) 🇭🇰. This repository aims to help students become familiar with speaker-related tasks while also serving as a resource for my own learning and development.

Summary of speaker related tasks, like speaker recognition, verification, diarization, spoofing, privacy, voice conversion, target speaker extraction and so on.

Book recommendations

• Understanding Deep learning [pdf]
• Computer vision: models learning and inference [pdf]
• 深入浅出强化学习：原理入门 [pdf]
• Reinforcement Learning [pdf]

Basic Knowledge of Machine Learning

• Kullback-Leibler Divergence
• Equal Error Rate
• minimum Detection Cost Function

Speaker Recognition/Verification:

Toolkit

• Wespeaker

Speaker Models

• overview:
  • Overview of Speaker Modeling and Its Applications: From the Lens of Deep Speaker Representation Learning
• i-vector
• d-vector (frame-level): Deep neural networks for small footprint textdependent speaker verification
• x-vector (segment-level):
  • X-vectors: Robust dnn embeddings for speaker recognition
  • Deep Neural Network Embeddings for Text-Independent Speaker Verification
  • ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification
• r-vector: But system description to voxceleb speaker recognition challenge 2019
• xi-vector: Xi-Vector Embedding for Speaker Recognition
• Transformer based:
  • Self Multi-Head Attention for Speaker Recognition
  • LOCAL INFORMATION MODELING WITH SELF-ATTENTION FOR SPEAKER VERIFICATION
• SPEAKER RECOGNITION FROM RAW WAVEFORM WITH SINCNET
• Self-supervised:
  • Self-supervised speaker embeddings
  • Neural Predictive Coding using Convolutional Neural Networks towards Unsupervised Learning of Speaker Characteristics
• PLDA: Probabilistic Linear Discriminant Analysis for Inferences About Identity
• Reshape Dimensions Network for Speaker Recognition
• Guided Speaker Embedding

Aggregation Layers

implementation: wespeaker/models/pooling_layers

• Temporal Average Pooling (TAP)
• Temporal Statistics Pooling (TSTP): X-vectors: Robust dnn embeddings for speaker recognition
• Attentive Statistics Pooling (ASP): ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification

Datasets

• Voxceleb1
• Voxceleb2

Challenge

• NIST SRE

Voice Conversion

Non-parallel:

• CycleGAN-VC: Non-parallel Voice Conversion Using Cycle-Consistent Adversarial Networks

Voice Anonymization

• MODELING PSEUDO-SPEAKER UNCERTAINTY IN VOICE ANONYMIZATION

Target Speaker Extraction

• INVESTIGATION OF SPEAKER REPRESENTATION FOR TARGET-SPEAKER SPEECH PROCESSING

Speaker Diarization

Spoofing

• Towards Quantifying and Reducing Language Mismatch Effects in Cross-Lingual Speech Anti-Spoofing

Targer Speaker ASR

Personalized VAD

Others

• Emerging Properties in Self-Supervised Vision Transformers