Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / clovaai/voxceleb_trainer issues and pull requests

#189 - softmaxproto loss for binary classification

Issue - State: open - Opened by ukemamaster 3 months ago

#188 - Audio sample distribution visualization

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

#187 - Multi node training

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

#186 - test/train_list format

Issue - State: closed - Opened by shreeshailgan 8 months ago

#184 - How can I get the USERNAME and PASSWORD?

Issue - State: closed - Opened by DiLiangWU 8 months ago - 7 comments

#182 - Missing Voxceleb1 Identification Dataset

Issue - State: open - Opened by Xiang47 8 months ago - 1 comment

#181 - test

Issue - State: closed - Opened by wwyl2000 8 months ago

#180 - Why the speed is very slow?

Issue - State: open - Opened by dimuthuanuraj 9 months ago - 3 comments

#178 - No Video Version of VoxCeleb1

Issue - State: closed - Opened by CloudWalkerW 11 months ago - 3 comments

#177 - Dataset cannot be downloaded

Issue - State: closed - Opened by zoroglu about 1 year ago - 3 comments

#176 - Dataset download

Issue - State: closed - Opened by yuelei0428 about 1 year ago - 7 comments

#175 - How to train used DINO framework

Issue - State: open - Opened by TingziGITHUB over 1 year ago - 1 comment

#174 - Training a binary classifier (in case of 2 speakers)

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

#173 - How to change the sample rate?

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

#172 - how to save a new pretrained model

Issue - State: closed - Opened by dntuong over 1 year ago - 1 comment

#171 - the url for data download can't work

Issue - State: closed - Opened by LeeHongji over 1 year ago - 7 comments

#170 - Issue with training Common Voice Dataset

Issue - State: closed - Opened by dimuthuanuraj over 1 year ago - 1 comment

#169 - Shouldn't Prototypical loss use torch.cdist instead of F.pairwise_distance?

Issue - State: open - Opened by theolepage almost 2 years ago - 2 comments
Labels: question

#168 - What validation data is using?

Issue - State: open - Opened by katsu-77 almost 2 years ago
Labels: question

#167 - Data preparation

Issue - State: closed - Opened by shiddiqie almost 2 years ago - 1 comment

#166 - Edit path in convert

Pull Request - State: closed - Opened by griko almost 2 years ago

#165 - A bug in convert function path

Issue - State: open - Opened by griko almost 2 years ago

#164 - ValueError: Conversion failed

Issue - State: closed - Opened by KajiMaCN almost 2 years ago - 1 comment

#163 - Data preparation

Issue - State: open - Opened by mattymz almost 2 years ago - 3 comments

#162 - CVE-2007-4559 Patch

Pull Request - State: closed - Opened by TrellixVulnTeam almost 2 years ago - 1 comment

#161 - Download Speed

Issue - State: closed - Opened by simasima121 about 2 years ago - 2 comments

#160 - Cannot achieve paper's RawNet3 results using an official recipe

Issue - State: open - Opened by happyjin about 2 years ago - 4 comments

#159 - Need new download script

Issue - State: open - Opened by brianw0924 about 2 years ago - 7 comments

#158 - training RawNet3 with new dataset

Issue - State: closed - Opened by angelabr over 2 years ago - 1 comment

#157 - Issues with pretrained model `RawNet3/model.pt`

Issue - State: closed - Opened by SangwonSUH over 2 years ago - 1 comment

#156 - assert x.size()[1] == self.in_feats AssertionError

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

#155 - Where is the DINO self-supervised framework?

Issue - State: closed - Opened by speaker-lover over 2 years ago - 3 comments

#154 - Add rawnet3, fix evaluation

Pull Request - State: closed - Opened by Jungjee over 2 years ago
Labels: bug, enhancement

#153 - GAN based losses.

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

#152 - Video Dataset of VoxCeleb2

Issue - State: closed - Opened by testzer0 over 2 years ago - 1 comment

#151 - Inquiries on multi-modal data loader

Issue - State: closed - Opened by ClaudiaShu over 2 years ago - 1 comment

#150 - Error in evaluateFromList() from SpeakerNet.py

Issue - State: closed - Opened by gems047 over 2 years ago - 1 comment

#149 - Question regarding dataset downloading

Issue - State: closed - Opened by itsmag11 over 2 years ago - 6 comments

#147 - The pretrained model of VGG

Issue - State: closed - Opened by muzihuole over 2 years ago - 1 comment

#146 - Question about angleproto.py code

Issue - State: closed - Opened by huangrunqian over 2 years ago - 3 comments

#145 - question on the pretrained models

Issue - State: closed - Opened by jkhourybbn over 2 years ago - 2 comments

#144 - Cannot get 1.1771 eer (a large gap) with the pretrained model in README

Issue - State: closed - Opened by cc98gg over 2 years ago - 12 comments

#143 - KeyError: 'id10004/JKMfqmjUwMU/00003.wav'

Issue - State: closed - Opened by VenkatesanSoundirarajan over 2 years ago - 1 comment

#142 - Can anyone help me with "Saving score file as follows?"

Issue - State: closed - Opened by dimuthuanuraj over 2 years ago - 2 comments

#141 - Can't unzip the folder

Issue - State: closed - Opened by ShayKaiserman over 2 years ago - 1 comment

#140 - chore: fix dependencies

Pull Request - State: closed - Opened by msh9184 almost 3 years ago

#139 - torch version for using 'all_gather_object'

Issue - State: closed - Opened by msh9184 almost 3 years ago - 2 comments

#138 - Use as Speaker Encoder for 48kHz audio

Issue - State: closed - Opened by astricks almost 3 years ago - 2 comments

#137 - How to calculate the test accuracy?

Issue - State: closed - Opened by dairuining almost 3 years ago - 1 comment

#136 - Reduce data waste with "resmixid"

Pull Request - State: open - Opened by IMLHF about 3 years ago

#135 - Is there anynone tried large margin finetune?

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

#134 - Short speech（< 1 sec）bad performance，how to improve？thanks!

Issue - State: closed - Opened by sundyll about 3 years ago - 1 comment

#133 - How to test with a different test list?

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

#132 - How you generate speaker embedding?

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

#131 - what is the mean TEER/TAcc?

Issue - State: closed - Opened by sundyll about 3 years ago - 2 comments

#130 - Different batch sizes will yield different performance results.

Issue - State: closed - Opened by rezimitpo about 3 years ago - 2 comments

#129 - The password and username

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

#128 - Is speech/voice activity detection a part of this implementation

Issue - State: closed - Opened by iiscleap about 3 years ago - 2 comments

#127 - Update download links

Pull Request - State: closed - Opened by ngoanpv about 3 years ago - 1 comment

#126 - Must the datasets convert to wav from m4a?

Issue - State: closed - Opened by sundyll about 3 years ago - 1 comment

#125 - training is extremely slow (single gpu v-100, ssd)

Issue - State: closed - Opened by zabir-nabil about 3 years ago - 2 comments

#124 - Download script has some error?

Issue - State: closed - Opened by ali2iptoki over 3 years ago - 4 comments

#123 - Lossfunction issue

Issue - State: closed - Opened by makimon123 over 3 years ago - 4 comments

#122 - Train list & Evaluation list

Issue - State: closed - Opened by dimuthuanuraj over 3 years ago - 2 comments

#121 - Difference between speakerNet and speakerNet of NIVIDIA?

Issue - State: closed - Opened by ali2iptoki over 3 years ago - 2 comments

#120 - how to regist the person voice and predict it by new wav file

Issue - State: closed - Opened by daizzhisheng over 3 years ago - 4 comments

#119 - SpeakerNet: score = -1 * numpy.mean(dist)

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

#118 - How to find the optimal threshold?

Issue - State: closed - Opened by ukemamaster over 3 years ago - 4 comments

#117 - dataset problem

Issue - State: closed - Opened by yaoyao1206 over 3 years ago - 4 comments

#116 - get error when use --mixedprec get

Issue - State: closed - Opened by jingxuan9862 over 3 years ago - 2 comments

#115 - import nsml

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

#113 - why use negtive as positive in triplet loss

Issue - State: closed - Opened by WilliamZhaoz over 3 years ago

#112 - triplet 等其他loss的training的code和命令求补充

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

#111 - Dataset download

Issue - State: closed - Opened by CaptainPrice12 over 3 years ago - 2 comments

#110 - Training slows down after few steps

Issue - State: closed - Opened by ukemamaster over 3 years ago - 2 comments

#109 - Issues when trying different loss functions

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

#108 - Training the baseline only

Issue - State: closed - Opened by ukemamaster over 3 years ago

#107 - difference between wavfile and soundfile?

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

#104 - Distributed

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

#100 - Similarity or distance metric to calculate scores?

Issue - State: closed - Opened by Nada-gh over 3 years ago - 2 comments

#99 - Pre-trained model files linked in the readme are not accessible anymore

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

#98 - Q: distribute training doesn't seems to split dataset between GPUs

Issue - State: closed - Opened by asimov-aiz over 3 years ago - 4 comments

#97 - About the changing training data sample rate

Issue - State: closed - Opened by whitegon over 3 years ago - 2 comments

#96 - about GRAPH ATTENTION NETWORKS FOR SPEAKER VERIFICATION

Issue - State: closed - Opened by seacj over 3 years ago

#95 - Question about converting to TensorFlow

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

#94 - questions about dimension transform

Issue - State: closed - Opened by forwiat almost 4 years ago - 1 comment

#93 - Bug with Multi gpu training

Issue - State: closed - Opened by 009deep almost 4 years ago - 8 comments

#92 - question about data transpose in SpeakerNet.py

Issue - State: closed - Opened by forwiat almost 4 years ago - 1 comment

#91 - Res2net model support

Pull Request - State: closed - Opened by stevenhillis almost 4 years ago - 2 comments

#90 - The configuration of HA4 in paper

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

#89 - Training difference between 2 work

Issue - State: closed - Opened by 009deep almost 4 years ago - 9 comments

#88 - Differences between the two implementations in paper [1] and voxsrc 20 summary paper

Issue - State: closed - Opened by Evan0-0 almost 4 years ago - 2 comments

#87 - Any recommendations on implementing TNDD using voxceleb_trainer?

Issue - State: closed - Opened by ShaneRun almost 4 years ago - 7 comments

#86 - Regarding ECAPA-TDNN implementation

Issue - State: closed - Opened by 009deep almost 4 years ago - 5 comments

#85 - Question about reading wav

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