google uis-rnn issues - Githubissues

google / uis-rnn

This is the library for the Unbounded Interleaved-State Recurrent Neural Network (UIS-RNN) algorithm, corresponding to the paper Fully Supervised Speaker Diarization.

https://arxiv.org/abs/1810.04719

Apache License 2.0

1.54k stars 319 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

assign gpu with arguments

#91 pthavarasa opened 1 year ago
0
Predicted labels doesn't match with Ground truth labels but the accuracy of test results is 0.8%

#90 SanaullahOfficial closed 1 year ago
1
[Bug] Making a prediction on CPU after training on GPU

#89 pthavarasa opened 1 year ago
0
Any documentations on training from scratch using custom data in other languages ?

#88 thangld201 opened 2 years ago
1
rnn initial state trainable

#87 jardnzm closed 2 years ago
1
Is there a way to fine tune an already existing pre-trained model?

#86 saumyaborwankar closed 3 years ago
1
[Question] The dimension of toy test data [test_sequence] is (25, 95, 256) what does the first 2 dimension represent? Toy train data [train_sequence] has dimension (4627, 256) which is understandable.

#85 saumyaborwankar closed 3 years ago
1
[Question] About num_non_zero

#84 shubhunegi30 closed 3 years ago
1
Is is possible to pre-load the model for multiple request?

#83 ghost closed 3 years ago
1
No module named coverage

#82 yerzhan7orazayev closed 3 years ago
1
Change input size

#81 surasakBoonkla closed 3 years ago
1
[Question] Are input d-vectors for training assumed L2-normalized?

#80 vadimkantorov closed 3 years ago
8
Unable to convert pytorch model to tensorflow in Diarization on mobile device.

#79 alamnasim closed 3 years ago
2
how to control the number of different speaker when predicting?

#78 NewEricWang closed 3 years ago
1
Fix typo: recommanded -> recommended

#77 mattmatters closed 3 years ago
2
uis-rnn gives different result on broken audios and continuous audios

#76 ashu170292 closed 3 years ago
5
Question about custom data generator

#75 YanaHontarenko opened 4 years ago
0
what is train data format?

#74 jkkj1630 closed 4 years ago
1
[Bug] Predict method does not finish

#73 ArlindKadra opened 4 years ago
3
about model

#72 zyc1310517843 closed 4 years ago
1
Embedding Extraction Procedure

#71 divyeshrajpura4114 closed 4 years ago
1
add crp_alpha support

#70 aluminumbox closed 4 years ago
8
I run the demo use toy data, found loss is negative number.

#69 Naminwang closed 4 years ago
1
Confusion about predicted labels

#68 clabornd closed 4 years ago
3
[Question]About UIS-RNN d-vector

#67 BarCodeReader closed 4 years ago
7
Repo

#66 chrisspen closed 4 years ago
1
Handling unknown speakers?

#65 chrisspen closed 4 years ago
1
Handling unknown speakers?

#64 chrisspen closed 4 years ago
3
uis-rnn error is too high.

#63 wychoi44 closed 4 years ago
1
How to create training data?

#62 chrisspen closed 4 years ago
2
[Question] Online Speaker Decentralize

#61 hungnvk54 closed 4 years ago
0
question about the sigma2 prior loss increasing during the training

#60 maxandchen closed 4 years ago
2
[Question] How can I get the text of each speaker after speaker diarization?

#59 rowhitswami closed 4 years ago
3
Loss calculation in prediction

#58 gen35 closed 4 years ago
2
corrected transition_bias estimation, fixes #55

#57 DonkeyShot21 closed 4 years ago
6
[Issue #26] add arg 'num_speaker'

#56 bencq closed 4 years ago
4
[Bug] Incorrect estimation of transition_bias

#55 DonkeyShot21 closed 4 years ago
13
Is UIS-RNN the technique used in Speaker Diarization in Google Speech-to-text API? [Question]

#54 blazejdolicki closed 4 years ago
1
Understanding the use of resize_sequence() and the batch creation for RNN training

#53 vickianand closed 4 years ago
2
[Question] how to relate sigma2_prior_loss() with the paper

#52 jprobichaud closed 5 years ago
1
Understanding diarization labels

#51 zyc1310517843 closed 5 years ago
6
uis-rnn can't work for long utterances dataset?

#50 wrongbattery opened 5 years ago
19
The number of speakers and whether the content of the speaker needs to be the same。

#49 zyc1310517843 closed 5 years ago
1
[Question] Which feature was used for VAD?

#48 seungwonpark closed 5 years ago
1
Any plan on upgrading to pytorch 1.0+?

#47 tanvirfuad closed 4 years ago
3
How to convert audio data into test data of algorithm for testing

#46 zyc1310517843 closed 5 years ago
1
[Question] Can I use speaker annotated datasets in other language rather than English?

#44 muntasir2000 closed 5 years ago
1
How to define a speaker per segments from overlapped widow of frames? Prediction on realtime data?

#43 alamnasim closed 5 years ago
5
Is the GRU really needed to predict mu_t ?

#42 hbredin closed 5 years ago
7
The clustering performance influenced by overlap window size

#41 taylorlu closed 5 years ago
7