Course project for EE698R (2020-21 Sem 2). An X-Vector Based Speaker Diarization System with AutoEncoder based clustering method. Also supports spectral and KMeans clustering method.
GPL-3.0 License
Signal processing and ML
Hierarchical Sketch Induction for Paraphrase Generation (Hosking et al., ACL 2022)
Hybrid Discriminative-Generative Training via Contrastive Learning
Generic image compressor for machine learning. Pytorch code for our paper "Lossy compression for ...
Vision-Augmented Retrieval and Generation (VARAG)
TristouNet: Triplet Loss for Speaker Turn Embedding
[ECCV2022] New benchmark for evaluating pre-trained model; New supervised contrastive learning fr...
本项目基于PaddleDetection目标检测开发套件,选取1.3M超轻量PPYOLO tiny进行项目开发,并部署于windows端。