# PercepNet(Under construct)
Unofficial implementation of PercepNet: A Perceptually-Motivated Approach for Low-Complexity, Real-Time Enhancement of Fullband Speech
https://www.researchgate.net/publication/343568932_A_Perceptually-Motivated_Approach_for_Low-Complexity_Real-Time_Enhancement_of_Fullband_Speech
## Todo
- [X] pitch estimation
- [X] Comb filter
- [X] ERBBand c++ implementation(need fix)
- [ ] Data Creater(c++) for pytorch
- [ ] DNNModel pytorch
- [ ] DNNModel c++ implementation
- [ ] Postfiltering
## SampleData
clean speech - VCTK 48k wav https://datashare.is.ed.ac.uk/handle/10283/2791 (clean_train_set)
noise data - DEMAND 48k wav https://zenodo.org/record/1227121#__sid=js0 (*.48k.zip)
# Reference
https://github.com/wil-j-wil/py_bank
https://github.com/dgaspari/pyrapt
https://github.com/xiph/rnnoise
PercepNet:（正在构建中）PercepNet的非官方实现
需积分: 50 128 浏览量
2021-05-18
21:44:25
上传
评论 1
收藏 7.72MB ZIP 举报
PercepNet-main.zip （29个子文件） 
PercepNet-main 
.gitignore 2KB
requirements.txt 13B src arch 2.13MB opus_types.h 5KB arch.h 7KB common.h 1KB kiss_fft.cpp 17KB kiss_fft.h 6KB erbband.h 3KB pitch.cpp 14KB pitch.h 5KB celt_lpc.cpp 7KB denoise.cpp 17KB _kiss_fft_guts.h 6KB rnnoise.h 2KB main.cpp 911B celt_lpc.h 2KB LICENSE 2KB utils pyrapt.py 33KB raptparams.py 2KB vorbis.py 0B vorbisstft.py 5KB __pycache__ filterbanks.cpython-35.pyc 5KB windows.py 1KB filterbanks.py 6KB sampledata noise noise.pcm 5.49MB speech speech.pcm 5.49MB README.md 830B rnn_train.py 1KB
评论0
最新资源