Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / mravanelli/SincNet issues and pull requests
#115 - Is the Learning Speaker Representations with Mutual Information paper ever going to be implemented?
Issue -
State: open - Opened by HannaSwida over 1 year ago
#114 - SincNet visualization
Issue -
State: open - Opened by ZaUt-bio over 1 year ago
#113 - cw_len
Issue -
State: open - Opened by yunzqq over 1 year ago
#112 - Clarifications for creating sequential batches instead of random sampling.
Issue -
State: open - Opened by Raviteja-banda almost 2 years ago
#111 - .scp files for VoxCeleb1/VoxCeleb2 - how to get/generate them
Issue -
State: open - Opened by HannaSwida almost 2 years ago
#110 - CUDA error: device-side assert triggered
Issue -
State: open - Opened by Raviteja-banda almost 2 years ago
#109 - Sincnet show the spectrogram is so odd ,it create so many stripe,why????
Issue -
State: open - Opened by nickthereal1997 almost 3 years ago
- 1 comment
#108 - Voxceleb1 cfg file?
Issue -
State: open - Opened by Tonmoy1321 about 3 years ago
#107 - Question about the creation of the Sinc filterbank
Issue -
State: open - Opened by fColangelo about 3 years ago
#106 - DownStream (ASV: Automatic Speaker Verification) - missing information
Issue -
State: closed - Opened by OG-algo over 3 years ago
#105 - fix pickle error when loading class_dict_file
Pull Request -
State: open - Opened by tejuafonja over 3 years ago
#104 - fix "list index out of range" error for fc_drop
Pull Request -
State: open - Opened by tejuafonja over 3 years ago
#103 - Need some idea on how to use this model to perform speaker verification task.
Issue -
State: closed - Opened by mosszhd over 3 years ago
#102 - loss function returns nan
Issue -
State: open - Opened by natank1 almost 4 years ago
#101 - the problem with the other dataset
Issue -
State: closed - Opened by zohrehansari about 4 years ago
#100 - Hamming window simplification
Issue -
State: open - Opened by vadimkantorov about 4 years ago
#99 - Difference between mfcc feature and torch tensor feature used while computing d vector? Does mfcc affect speaker identification accuracy?
Issue -
State: open - Opened by alamnasim about 4 years ago
#98 - Can it work on pure music?
Issue -
State: closed - Opened by lerogo about 4 years ago
#97 - SincNet band
Issue -
State: open - Opened by piplyman about 4 years ago
#96 - Use of two optimizers?
Issue -
State: open - Opened by lucvanwyk about 4 years ago
#95 - chunk numbers in code "N_fr"
Issue -
State: open - Opened by brealisty about 4 years ago
#94 - How long did it take to run the Timit dataset?
Issue -
State: open - Opened by JucyCherry about 4 years ago
- 1 comment
#93 - Librispeech as used in the paper
Issue -
State: closed - Opened by alkaet over 4 years ago
#92 - What "compute_d_vectors.py" is really about?
Issue -
State: open - Opened by briverse17 over 4 years ago
- 1 comment
#91 - How to quickly train the model based on the original model for the audio data of the new speaker?
Issue -
State: open - Opened by joewale over 4 years ago
- 1 comment
#90 - Is this a bug?
Issue -
State: open - Opened by JunzheJosephZhu over 4 years ago
- 1 comment
#89 - Gradient backward of the first filter is much lower than other layers.
Issue -
State: open - Opened by mechanicalsea over 4 years ago
#88 - Cannot replicate the cumulative frequency response of the SincNet filters
Issue -
State: closed - Opened by tuananh0305 over 4 years ago
- 3 comments
#87 - Shift vs overlap
Issue -
State: open - Opened by JunzheJosephZhu over 4 years ago
- 2 comments
#86 - The error when run speaker_id.py
Issue -
State: open - Opened by BLack-yzf over 4 years ago
- 1 comment
#85 - Error while generating .npy file
Issue -
State: open - Opened by bsshruthi22 almost 5 years ago
#84 - Cannot reproduce cumulative frequency response of SincNet on Speaker-id
Issue -
State: open - Opened by thibaultallenet-cea almost 5 years ago
- 4 comments
#83 - About eq.4 part in sincConv_fast code
Issue -
State: closed - Opened by alxiom almost 5 years ago
- 1 comment
#82 - The note about self.n_ is wrong
Issue -
State: closed - Opened by TanTingyi almost 5 years ago
- 1 comment
#81 - Amplification factor
Issue -
State: closed - Opened by CarmiShimon almost 5 years ago
- 2 comments
#80 - Librispeech dataset - version used in paper
Issue -
State: closed - Opened by ghost almost 5 years ago
#79 - Frequency response of sinc filters
Issue -
State: closed - Opened by zehaitu almost 5 years ago
- 5 comments
#78 - Musical instruments classification
Issue -
State: closed - Opened by CarmiShimon almost 5 years ago
- 1 comment
#77 - EER is very high on dataset VoxCeleb1
Issue -
State: open - Opened by heyayun18188 almost 5 years ago
- 2 comments
#76 - SincConv on multiple channels
Issue -
State: closed - Opened by chanshing almost 5 years ago
- 1 comment
#75 - training loss is 'nan'
Issue -
State: closed - Opened by 008karan almost 5 years ago
- 4 comments
#74 - SincNet vs. STFT
Issue -
State: closed - Opened by snakers4 almost 5 years ago
- 26 comments
#73 - TIMIT data download
Issue -
State: closed - Opened by heyayun18188 almost 5 years ago
- 1 comment
#72 - Any class detector
Issue -
State: closed - Opened by CarmiShimon almost 5 years ago
- 3 comments
#71 - Big data loading - time consumption
Issue -
State: closed - Opened by CarmiShimon about 5 years ago
- 2 comments
#70 - Code reformat, refactor and bug fixes
Pull Request -
State: open - Opened by qwfy about 5 years ago
#69 - Cumulative frequency response of the SincNet filters
Issue -
State: closed - Opened by javi-99 about 5 years ago
- 2 comments
#68 - Performance on speaker identification is good ,but on speaker verification is poor
Issue -
State: open - Opened by zrtlemontree about 5 years ago
- 4 comments
#67 - Performance on emotion classification is poor?
Issue -
State: closed - Opened by wrongbattery about 5 years ago
- 4 comments
#66 - performance gap between the pretrained model and that in the paper on TIMIT dataset
Issue -
State: open - Opened by smallflyingpig about 5 years ago
- 5 comments
#65 - predict the speaker of a file
Issue -
State: closed - Opened by gerardgilliland about 5 years ago
- 2 comments
#64 - Create to_upper.py
Pull Request -
State: open - Opened by yunfei-shi about 5 years ago
#63 - pt_file error
Issue -
State: closed - Opened by gerardgilliland about 5 years ago
- 6 comments
#62 - Is the offset-by-one error on purpose?
Issue -
State: open - Opened by qwfy about 5 years ago
#61 - What's the reason of `Batch_dev = N_fr` when `avoid_small_en_fr = True`?
Issue -
State: closed - Opened by qwfy about 5 years ago
- 2 comments
#60 - low_hz frequency
Issue -
State: closed - Opened by Mistobaan about 5 years ago
- 1 comment
#59 - Computational Efficiency Through Symmetric Convolutions
Issue -
State: closed - Opened by KnurpsBram over 5 years ago
- 2 comments
#58 - Correct normalization
Pull Request -
State: closed - Opened by vroger11 over 5 years ago
- 1 comment
#57 - Problem on SER %
Issue -
State: closed - Opened by Sathiyakugan over 5 years ago
- 1 comment
#56 - performance on Voxceleb1?
Issue -
State: closed - Opened by linjucs over 5 years ago
- 6 comments
#55 - Removing non-speech intervals
Issue -
State: closed - Opened by bwang482 over 5 years ago
- 2 comments
#54 - speaker verification with GE2E loss
Issue -
State: closed - Opened by wrongbattery over 5 years ago
- 9 comments
#53 - How to divide the test set?
Issue -
State: closed - Opened by Range0122 over 5 years ago
- 6 comments
#52 - Update speaker_id.py
Pull Request -
State: open - Opened by rickychanhoyin over 5 years ago
#51 - How to do the GPU parallel with this code?
Issue -
State: closed - Opened by DanielMengLiu over 5 years ago
- 2 comments
#50 - TypeError: ufunc 'add' did not contain a loop with signature matching types dtype('<U32') dtype('<U32') dtype('<U32')
Issue -
State: closed - Opened by Archan2607 over 5 years ago
- 3 comments
#49 - computed d-vectors aren't consistent across runs
Issue -
State: closed - Opened by jprobichaud over 5 years ago
- 1 comment
#48 - Update speaker_id.py
Pull Request -
State: closed - Opened by rickychanhoyin over 5 years ago
- 1 comment
#47 - Configuration
Issue -
State: closed - Opened by gerardgilliland over 5 years ago
#46 - training on new datset
Issue -
State: closed - Opened by akashicMarga over 5 years ago
- 10 comments
#45 - error in create_batches
Issue -
State: closed - Opened by oualido1212 over 5 years ago
- 5 comments
#44 - TIMIT_preparation.py
Issue -
State: closed - Opened by harishmalaiappan over 5 years ago
#43 - TIMIT_preparation.py
Issue -
State: closed - Opened by harishmalaiappan over 5 years ago
- 1 comment
#42 - TIMIT (.wrd, .txt, .phn) file interpretations (numbers in front of the line)
Issue -
State: closed - Opened by andriuslauraitis over 5 years ago
- 4 comments
#41 - Question about TIMIT_labels.npy and others
Issue -
State: closed - Opened by andriuslauraitis over 5 years ago
- 3 comments
#40 - Installation Update
Issue -
State: closed - Opened by gerardgilliland over 5 years ago
- 2 comments
#39 - Taking the whole speech sequence as input without chunking
Issue -
State: closed - Opened by bwang482 over 5 years ago
- 11 comments
#38 - Re-training the trained SincNet on new dataset
Issue -
State: closed - Opened by hdubey over 5 years ago
- 11 comments
#37 - Adding new speakers/using transfer learning
Issue -
State: closed - Opened by bogdan-nechita over 5 years ago
- 8 comments
#36 - feat: add MIT License file
Pull Request -
State: closed - Opened by hbredin over 5 years ago
- 3 comments
#35 - Curious about performance of multiple optimizers vs single optimizer
Issue -
State: closed - Opened by seandickert over 5 years ago
- 3 comments
#34 - Training on GPU got error on CPU decoding
Issue -
State: closed - Opened by hdubey over 5 years ago
- 1 comment
#33 - Where SincNet is implemented?
Issue -
State: closed - Opened by maomaoqiuqiu over 5 years ago
- 1 comment
#32 - trained model error
Issue -
State: closed - Opened by taylorlu over 5 years ago
- 4 comments
#31 - how to download pretrained model??
Issue -
State: closed - Opened by ghost over 5 years ago
- 4 comments
#30 - AttributeError: module 'torch' has no attribute 'flip'
Issue -
State: closed - Opened by ravikirancap over 5 years ago
- 4 comments
#29 - How can I used sincnet for speaker verification task?
Issue -
State: closed - Opened by bharatsubedi41 over 5 years ago
- 8 comments
#28 - Can SincNet be used for speech emotion recognition?
Issue -
State: closed - Opened by maomaoqiuqiu over 5 years ago
- 4 comments
#27 - Hello, can SincNet extract features from the spectrograph?
Issue -
State: closed - Opened by maomaoqiuqiu over 5 years ago
#26 - add missing default sample rate
Pull Request -
State: closed - Opened by seungwonpark over 5 years ago
- 1 comment
#25 - voxCeleb1 and libri speech
Issue -
State: closed - Opened by hdubey over 5 years ago
- 15 comments
#24 - Rationale for dividing speech into chunks of 200ms with 10ms overlap
Issue -
State: closed - Opened by vikigenius over 5 years ago
- 3 comments
#23 - Perform speaker identification
Issue -
State: closed - Opened by anicolson over 5 years ago
- 16 comments
#22 - Run TIMIT data preparation.
Issue -
State: closed - Opened by maomaoqiuqiu over 5 years ago
- 11 comments
#21 - is it possible for passes x-vectors into sincnet and classifying a speaker. like speaker verification.?
Issue -
State: closed - Opened by MuruganR96 over 5 years ago
- 2 comments
#20 - SincNet Weights
Issue -
State: closed - Opened by Mohit17028 over 5 years ago
- 1 comment
#19 - did it had mutual Information to the open code?
Issue -
State: closed - Opened by eeric over 5 years ago
- 3 comments
#18 - Improving SincNet results on TIMIT by adding reverberation
Issue -
State: closed - Opened by seandickert over 5 years ago
- 5 comments
#17 - Reproducing LibriSpeech results
Issue -
State: closed - Opened by seandickert over 5 years ago
- 4 comments
#16 - Is sincnet possible to convert back to original wave?
Issue -
State: closed - Opened by r06944010 over 5 years ago
- 3 comments