ishine's Projects
一个推理库的实现, A DIY deep learning inference framework.
End-to-end spoken keyword search using Convolutional Neural Networks.
Attention-based model for keywords spotting
Mining effective negative training samples for keyword spotting (PyTorch)
Keyword spotting, Speech wake_up, pytorch, DNN, CNN, TDNN, DFSMN, LSTM
A decoder for finite state models for text processing.
Label Studio is a multi-type data labeling and annotation tool with standardized output format
:metal: LabelImg is a graphical image annotation tool and label object bounding boxes in images
个人实现的基于Django与semantic-ui的语言计算实验平台, 功能包括自然语言综合处理,词语计算,社会热点计算,人物计算,文学画像,职位画像等社会计算功能
End to end text to speech system using gruut and onnx
A fast, local neural text to speech system
This is the TensorFlow implementation of the Google LAS model.
Language-Agnostic SEntence Representations
This repo hosts the code and model of "Separate What You Describe: Language-Queried Audio Source Separation".
Lattice combination algorithm to combine inaccurate transcripts with hypothesis lattices
Bi-directional Lattice Recurrent Neural Networks for Confidence Estimation
Library to scrape and clean web pages to create massive datasets.
Official implementation of the paper: "LDNet: Unified Listener Dependent Modeling in MOS Prediction for Synthetic Speech"
largest-ever Automatic Speech Recognition leaderboard, periodically benchmarks SOTA commercial ASR APIs from Alibaba, Baidu, Google, IFlytek, Microsoft and so on.
Diffusion and Mutual Information-Based Target Speaker SVS by Learning from Singing Teacher
Pytorch implementation of LearnableUpsamplingLayer (NaturalSpeech, Tan et al., 2022)