xiph/rnnoise issues and pull requests

#241 - max msp external

Issue - State: open - Opened by trummerschlunk about 1 month ago

#240 - Training models

Issue - State: open - Opened by SolarGuy8 about 1 month ago

#239 - fixes for non-avx targets

Pull Request - State: open - Opened by marler8997 about 2 months ago

#238 - what are the changes to be made if we want to change the FRAME SIZE from 480 to 160?

Issue - State: open - Opened by sanjuktasr 2 months ago

#236 - what are the changes to be made if we want too change the FRAME SIZE from 480 to 160?

Issue - State: open - Opened by sanjuktasr 3 months ago - 1 comment

#235 - How to convert the model to rnnoise_data_little.c

Issue - State: open - Opened by AotYan 3 months ago - 1 comment

#234 - Update vec_neon.h

Pull Request - State: open - Opened by avcodecs 3 months ago

#233 - Training with already Clean and Noisy Data

Issue - State: closed - Opened by rosotron 3 months ago - 2 comments

#232 - Almagamation of rnnoise for having it single-source

Pull Request - State: closed - Opened by ValentiWorkLearning 4 months ago

#231 - Comparison of lite vs full model in 0.2.0

Issue - State: open - Opened by ggarber 5 months ago - 1 comment

#230 - Enable Discussions feature and convert "issues" that are not issues to a discussion

Issue - State: open - Opened by johncf 5 months ago

#229 - Control the size of model

Issue - State: closed - Opened by BancoLin 5 months ago - 1 comment

#228 - Fix SIMD fallback code when AVX2 is not available

Pull Request - State: open - Opened by j-schultz 5 months ago

#227 - CMake support, modernized rnnoise example with libsoundfile and CI setup based on Github Actions

Pull Request - State: open - Opened by ValentiWorkLearning 6 months ago - 2 comments

#226 - Callback implementation for xcorr kernel offload

Pull Request - State: closed - Opened by ValentiWorkLearning 6 months ago

#225 - reference values of training loss

Issue - State: open - Opened by BancoLin 6 months ago

#224 - Why is a biquad filter needed before denoise?

Issue - State: closed - Opened by segu-L 6 months ago - 1 comment

#223 - Possible to support basic (non-VSX) Altivec?

Issue - State: open - Opened by barracuda156 7 months ago

#222 - RNNoise v0.2 does not compile on Mac and Linux

Issue - State: open - Opened by grazder 7 months ago - 8 comments

#221 - Releasing RNNoise version 0.2

Issue - State: open - Opened by jmvalin 7 months ago - 15 comments

#220 - New model and default branch has changed from "master" to "main"

Issue - State: open - Opened by jmvalin 7 months ago - 4 comments

#219 - handle non-vla compilers

Pull Request - State: closed - Opened by marler8997 12 months ago - 9 comments

#218 - About rnnoise_demo output

Issue - State: open - Opened by christiancelona about 1 year ago - 1 comment

#215 - Data imbalance problem for Voice Activity Detection output block

Issue - State: open - Opened by bisnu16cse039 over 1 year ago

#214 - file generated unreadable

Issue - State: open - Opened by victorrosi over 1 year ago - 3 comments

#213 - rnnoise not filtering this specific noise

Issue - State: open - Opened by Ryuo1 over 1 year ago

#212 - About memory

Issue - State: open - Opened by panhu almost 2 years ago - 2 comments

#211 - What is the range of the input data?

Issue - State: closed - Opened by nosleeve1 about 2 years ago

#210 - Fix Wstrict-prototypes warnings

Pull Request - State: closed - Opened by JohannesKauffmann about 2 years ago - 1 comment

#209 - Potential underflow / overflow when converting from float to int16

Issue - State: open - Opened by Meoo over 2 years ago - 1 comment

#208 - Questions About Training

Issue - State: closed - Opened by ZihCode over 2 years ago

#207 - About the result

Issue - State: open - Opened by panhu over 2 years ago

#206 - Garbage Collector

Pull Request - State: closed - Opened by CrusaderDeleters over 2 years ago

#205 - Replace bad manual traps (error in modern compiler vers)

Pull Request - State: closed - Opened by wallabra over 2 years ago - 3 comments

#204 - Noise data

Issue - State: open - Opened by FanhuaandLuomu over 2 years ago

#203 - Is model accuracy affected by buffer size of scriptProcessor node for js implementation?

Issue - State: open - Opened by A-KVerma over 2 years ago - 1 comment

#202 - Hey, Congratulations for the awesome work. I was testing the code and seems like buffer size used while creating the script Processor node is impacting the model accuracy. For Ex:

Issue - State: closed - Opened by A-KVerma over 2 years ago

#201 - rnnoise eliminates the speaking voice

Issue - State: open - Opened by wa-heng almost 3 years ago

#200 - Noise at the start of every processed audio

Issue - State: open - Opened by Domhnall-Liopa almost 3 years ago - 3 comments

#199 - confused with pitch filter

Issue - State: open - Opened by GreenHandLW almost 3 years ago

#198 - about rnn_train.py

Issue - State: open - Opened by miaoyatai about 3 years ago

#197 - how to compile noise.js from trained weights

Issue - State: open - Opened by vishaldhull09 about 3 years ago

#196 - How to train with large dataset

Issue - State: open - Opened by Bach1502 about 3 years ago - 5 comments

#195 - Compile this into .DLL

Issue - State: closed - Opened by CountDookie about 3 years ago

#194 - procedure for fine tuning model instead of training from scratch

Issue - State: open - Opened by vishaldhull09 about 3 years ago

#193 - Use std::vector

Issue - State: closed - Opened by WindowsNT about 3 years ago - 5 comments

#192 - using RNNoise in browser

Issue - State: open - Opened by vishaldhull09 over 3 years ago - 2 comments

#191 - Create SIMD-accelerated version of `compute_gru` function

Pull Request - State: open - Opened by Ameobea over 3 years ago - 3 comments

#190 - Update rnn.c, compute_gru support Keras double bias

Pull Request - State: open - Opened by cookcodes over 3 years ago

#189 - Training on MS-SNSD dataset.

Issue - State: closed - Opened by Vatsalparsaniya over 3 years ago - 14 comments

#188 - Basic cmake support

Pull Request - State: open - Opened by Dmitry-Nikishov over 3 years ago

#187 - zero samples...

Issue - State: open - Opened by simon-cas over 3 years ago - 2 comments

#186 - Can't open rnnoise-master/src/denoise_training

Issue - State: open - Opened by RXAldreezee over 3 years ago

#185 - Why haven’t you used TensorFlow lite to use your RNNoise model in Android ? Why have you ported into C

Issue - State: open - Opened by Khubaib96 over 3 years ago - 1 comment

#184 - RNNoise not doing anything, result is identical to original

Issue - State: closed - Opened by DatCaptainHorse over 3 years ago - 6 comments

#183 - Update README (Added specific instructions for mac users)

Pull Request - State: open - Opened by malhotrachetan over 3 years ago

#182 - Fix a memory leak in denoise.c

Pull Request - State: closed - Opened by a-rose over 3 years ago - 2 comments

#181 - Add .gitignore

Pull Request - State: open - Opened by a-rose over 3 years ago

#180 - Understanding the pitch filter

Issue - State: open - Opened by davlar1 over 3 years ago

#179 - UnboundLocalError: local variable 'batch_index' referenced before assignment

Issue - State: open - Opened by ifishtoo over 3 years ago

#178 - How can I use it to add noise suppresion to a file?

Issue - State: open - Opened by DVDTSB over 3 years ago

#177 - Why the testing code only use half of the size of gru bias

Issue - State: closed - Opened by benlin1996 over 3 years ago - 1 comment

#176 - Question About generate_model ValueError

Issue - State: open - Opened by aron566 over 3 years ago

#175 - How to prepare training data

Issue - State: open - Opened by jacky-ge over 3 years ago

#174 - Test

Pull Request - State: closed - Opened by tomleavy over 3 years ago

#173 - Tl scanoss

Pull Request - State: closed - Opened by tomleavy over 3 years ago

#172 - Does anyone tried to implement this in Raspberry Pi?

Issue - State: open - Opened by RXAldreezee over 3 years ago - 3 comments

#171 - Question about dump_rnn.py

Issue - State: open - Opened by benlin1996 over 3 years ago

#170 - What is the noise_train used for

Issue - State: open - Opened by benlin1996 over 3 years ago

#169 - different length of .wav file after denoise

Issue - State: open - Opened by benlin1996 over 3 years ago - 4 comments

#168 - Code for mixing the origin signal with noise

Issue - State: open - Opened by benlin1996 over 3 years ago - 2 comments

#165 - Clashing symbols with Opus

Issue - State: open - Opened by Krzmbrzl over 3 years ago - 5 comments

#164 - Add USE_MALLOC define, to toggle the use of malloc() and free() instead of VLA

Pull Request - State: open - Opened by davidebeatrici over 3 years ago - 6 comments

#162 - Support for ffmpeg arnndn filter?

Issue - State: open - Opened by sarek9 almost 4 years ago - 31 comments

#161 - Train rnn with microsoft SNSD

Pull Request - State: open - Opened by jiangjianping almost 4 years ago

#155 - Output file generated using demo code is not playable

Issue - State: closed - Opened by rawatnaresh almost 4 years ago - 4 comments

#153 - Can this be used for pulseeffects?

Issue - State: closed - Opened by MikeyBeez almost 4 years ago - 2 comments

#152 - 5000000*87, what does (87-42) mean?

Issue - State: open - Opened by maggie0830 almost 4 years ago - 3 comments

#148 - How to evaluate the performance of denoising when i Compared two weights?

Issue - State: open - Opened by wangohaha almost 4 years ago - 4 comments

#145 - A question about the adaptation for different frame size e.g. 10ms 22band

Issue - State: open - Opened by 20050710212 about 4 years ago - 3 comments

#137 - Please tag releases

Issue - State: open - Opened by darix about 4 years ago - 7 comments

#119 - How to make this work real-time

Issue - State: open - Opened by viral98 over 4 years ago - 2 comments

#111 - error when make

Issue - State: open - Opened by HuongNV96 over 4 years ago - 2 comments

#106 - Not being maintained? I want to take care of it.

Issue - State: open - Opened by thisishoon almost 5 years ago - 5 comments

#100 - Integrating with OSS-Fuzz

Issue - State: closed - Opened by Google-Autofuzz almost 5 years ago - 1 comment

#99 - Has anyone built this and tested the demo in Windows>

Issue - State: open - Opened by sumanrbt almost 5 years ago - 7 comments
Labels: question

#98 - How to get speech only data-set used in this project?

Issue - State: open - Opened by ShintaZachariah about 5 years ago - 1 comment
Labels: question

#97 - How to train rrnoise with multiple files at a time?

Issue - State: open - Opened by ShintaZachariah about 5 years ago - 1 comment
Labels: question

#96 - What is the role of FRAME_SIZE_SHIFT?

Issue - State: open - Opened by aliblieli about 5 years ago
Labels: question

#95 - why training.f32's size is 43550089 ? it should be 43500000!

Issue - State: open - Opened by NileZhou about 5 years ago - 2 comments

#94 - Error while compiling

Issue - State: open - Opened by gianpaolocoro about 5 years ago

#93 - Is it possible to change supported FRAME_SIZE?

Issue - State: open - Opened by Ornolfr about 5 years ago - 3 comments

#92 - How to train and run the network with audio with a sampling rate of 16k

Issue - State: open - Opened by LikeSwim about 5 years ago - 3 comments
Labels: question

#91 - How about this codes in RNN???

Issue - State: open - Opened by mangwaier about 5 years ago - 3 comments

#90 - Installation Issue

Issue - State: closed - Opened by ghost about 5 years ago - 3 comments

#89 - White Noise issue.

Issue - State: open - Opened by sporwar-lifesize over 5 years ago - 2 comments
Labels: question

#88 - Windows and CMake support

Pull Request - State: open - Opened by almoghamdani over 5 years ago - 9 comments

#87 - who has the Loss Logs in training ?

Issue - State: open - Opened by ucasiggcas over 5 years ago - 1 comment
Labels: question

#86 - how to test the model after dump_rnn.py

Issue - State: open - Opened by ucasiggcas over 5 years ago
Labels: question

#85 - bug in dump_rnn.py

Issue - State: open - Opened by ucasiggcas over 5 years ago - 9 comments

GitHub / xiph/rnnoise issues and pull requests