Adversarial-domain-adaptation

This project is about domain adaptation in speech by adversarial network and deep neural naetwork.

My codes are Completed or in progress :

1. Speaker-Verification-Wav2Vec2-ECAPA-TDNN-Voxceleb1-Deepmine.ipynb (Completed)
   
2. Part1.ipynb (In progress)
   

links :

1. HF course:https://huggingface.co/course/chapter2/5?fw=pt
   
2. Speech brain github: https://github.com/speechbrain/speechbrain
   
3. pytorch : https://www.learnpytorch.io/00_pytorch_fundamentals/
   
4. baseline code : https://github.com/facebookresearch/fairseq
   
5. baseline article : https://arxiv.org/abs/2110.05777
   
6. asr like baseline : https://arxiv.org/abs/2110.00165
   
7. result of baseline (EER in Speaker Verification) : https://github.com/microsoft/UniSpeech
   new article can be the base and heppingÖ
8. Meta-Generalization for Domain-Invariant Speaker Verification
   
9. MFA-Conformer: Multi-scale Feature Aggregation Conformer for Automatic Speaker Verification
   
10. LARGE-SCALE SELF-SUPERVISED SPEECH REPRESENTATION LEARNING FOR AUTOMATIC SPEAKER VERIFICATION-my base1
    
11. End-to-end speaker segmentation for overlap-aware resegmentation
    
12. Large-scale Self-Supervised Speech Representation Learning for Automatic Speaker Verification
    

Implementation's Hint :
0. https://colab.research.google.com/github/m3hrdadfi/notebooks/blob/main/Fine_Tune_XLSR_Wav2Vec2_on_Persian_ShEMO_ASR_with_%F0%9F%A4%97_Transformers_ipynb.ipynb (very important

1. https://huggingface.co/microsoft/wavlm-base-sv
   
2. https://huggingface.co/spaces/microsoft/wavlm-speaker-verification/blob/main/app.py
   
3. https://huggingface.co/speechbrain/spkrec-ecapa-voxceleb
   
4. https://huggingface.co/docs/transformers/model_doc/xlsr_wav2vec2 (Left PART IS IMPORTANT)
   
5. https://huggingface.co/datasets/superb