Discover @ZhihaoDU Open Source projects

Zhihao Du (@ZhihaoDU)

ZhihaoDU

Stars
128
Global Rank 172,498 (Top 6 %)
Followers 45
Following 7
Registered over 7 years ago
Most used languages

Python
80.0 %

HTML
20.0 %
Location 🇨🇳 China
Country Total Rank 19,818
Country Ranking

Python
4,015

speech_feature_extractor

Some useful features of speech process, such as MFCC, gammatone filterbank, GFCC, spectrum(power spectrum and log-power spectrum), Amplitude Modulation Spectrum(AMS) and so on.

du2022sond

Speaker overlap-aware Neural Diarization

du2020dan

The implementation of our paper "Double adversarial networks for monaural speech enhancement" accepted by INTERSPEECH 2020.

food_is_unstopped

Food is unstopped!!!! GO!

du2020kws

This a small footprint robust KWS system which is based on the multi-conditional training, retraining and joint-training. This system includes a small-footprint KWS system and a small-footprint speech enhancement model. We also investigate a compress method for CNN and LSTM.

zhihaodu.github.io