All Projects → Turing311 → Realtime_AudioDenoise_EchoCancellation

Turing311 / Realtime_AudioDenoise_EchoCancellation

Licence: other
No description or website provided.

Programming Languages

C++
36643 projects - #6 most used programming language
c
50402 projects - #5 most used programming language
objective c
16641 projects - #2 most used programming language

Projects that are alternatives of or similar to Realtime AudioDenoise EchoCancellation

speexdsp-python
Speex Echo Canceller Python Library
Stars: ✭ 88 (-3.3%)
Mutual labels:  speex, aec
DDCToolbox
Create and edit DDC headset correction files
Stars: ✭ 70 (-23.08%)
Mutual labels:  audio-processing
R8brain Free Src
High-quality pro audio sample rate converter / resampler C++ library
Stars: ✭ 238 (+161.54%)
Mutual labels:  audio-processing
MusicVisualizer
A music visualizer based on the ATMEGA328P-AU
Stars: ✭ 30 (-67.03%)
Mutual labels:  audio-processing
simple-waveform-visualizer
JS Audio API 놀이터
Stars: ✭ 31 (-65.93%)
Mutual labels:  audio-processing
emusic net
Neural network to classify certain styles of Electronic music
Stars: ✭ 22 (-75.82%)
Mutual labels:  audio-processing
Otto
Sampler, Sequencer, Multi-engine synth and effects - in a box! [WIP]
Stars: ✭ 2,390 (+2526.37%)
Mutual labels:  audio-processing
SpleeterRT
Real time monaural source separation base on fully convolutional neural network operates on Time-frequency domain.
Stars: ✭ 111 (+21.98%)
Mutual labels:  audio-processing
Sound-based-bird-species-detection
Sound-based Bird Classification - using AI, acoustics and ornithology to classify birds in the environment, an environmental awareness project (Web Application, Flask, Python)
Stars: ✭ 56 (-38.46%)
Mutual labels:  audio-processing
speckle-blender
speckle.systems/tag/blender/
Stars: ✭ 38 (-58.24%)
Mutual labels:  aec
ACA-Slides
Slides and Code for "An Introduction to Audio Content Analysis," also taught at Georgia Tech as MUSI-6201. This introductory course on Music Information Retrieval is based on the text book "An Introduction to Audio Content Analysis", Wiley 2012/2022
Stars: ✭ 84 (-7.69%)
Mutual labels:  audio-processing
ISP-pipeline-hdrplus
Denoise,HDR,Isppipeline,Image-processing(图形处理),camera, Isp ,HDRplus
Stars: ✭ 412 (+352.75%)
Mutual labels:  denoise
sonopy
A simple audio feature extraction library
Stars: ✭ 72 (-20.88%)
Mutual labels:  audio-processing
vim-webgl-viewer
An easy to use online 3D model viewer based on Three.JS.
Stars: ✭ 41 (-54.95%)
Mutual labels:  aec
keras-audio
keras project for audio deep learning
Stars: ✭ 37 (-59.34%)
Mutual labels:  audio-processing
Fourier
Fast Fourier transforms (FFTs) in Rust
Stars: ✭ 206 (+126.37%)
Mutual labels:  audio-processing
android-vad
This VAD library can process audio in real-time utilizing GMM which helps identify presence of human speech in an audio sample that contains a mixture of speech and noise.
Stars: ✭ 64 (-29.67%)
Mutual labels:  audio-processing
web-voice-processor
A library for real-time voice processing in web browsers
Stars: ✭ 69 (-24.18%)
Mutual labels:  audio-processing
Tensorflow-Audio-Classification
Audio classification with VGGish as feature extractor in TensorFlow
Stars: ✭ 105 (+15.38%)
Mutual labels:  audio-processing
FluX
A convenient way of processing digital signals in F#
Stars: ✭ 17 (-81.32%)
Mutual labels:  audio-processing

Realtime_AudioDenoise_EchoCancellation

This project is c++ port of DTLN denoise.
For denoise and AEC purpose, I've tried speex, webrtc, rnnoise and etc, but didn't find optimal solution (realtime & performance).
Lately, I found DTLN solution by breizhn.
DTLN not only performs denoising, but also has deecho effect.

It's VC++ project, but use pure c++, so you can adopt to any platforms.
I've included all the dependencies for wav, tflite, fft, resampling.
Also I've included pretrained tflite models from DTLN project.
So you can test project using raw pcm or wav files in various format.

USAGE

You can create multiple instances.
Please refer to project for detailed usage.
Input format of denoise function is 16KHz, fp32 pcm.

void tflite_create(trg_engine* engine)
void trg_denoise(trg_engine* engine, float* samples, float* out, int sampleCount)
void tflite_destroy(trg_engine* engine)

TIPS

I've tested project on armv7 1GHz environment and got realtime performance.
If performance not match your need, you can use quantized model in models folder and tried to input 8Khz directly instead of 16Khz pcm**.
** I've tested 8Khz pcm directly without resampling and got x2 speed up, reasonable quality on test samples. (Not recommended)

Personally thanks to breizhn for his great work

And looking forward to DTLN-aec repo completion.

STAR if project is helpful

And welcome pull requests and issue.

REFERENCES

DTLN (https://github.com/breizhn/DTLN) (Great thanks to breizhn)
PFFT (https://gitlab.mpcdf.mpg.de/mtr/pocketfft/-/tree/cpp)
Resample (https://github.com/cpuimage/resampler)

LICENSE

Anti 996 License from (https://github.com/kattgu7/Anti-996-License)

2021.04.11

Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].