google/uis-rnn

Issues

• Add a `online_predict()` API for streaming input

  #28 opened 6 years ago by wq2012
  3

• Batch prediction? - or allow prediction using multiprocessing

  #32 opened 5 years ago by hbredin
  12

• Question on the generative process of UIS-RNN

  #20 opened 5 years ago by 123456789077
  3

• about the training loss and the batch size

  #33 opened 5 years ago by simpleishappy
  8

• Add support for estimation of crp_alpha

  #4 opened 6 years ago by wq2012
  5

• handle overlapped speech

  #18 opened 6 years ago by PES2g
  3

• Performance degrade for multi-person meeting

  #29 opened 6 years ago by PES2g
  3

• [Question] How to prepare embedding data for training UIS-RNN?

  #37 opened 6 years ago by OpenCVnoob
  5

• Consider making the package available in conda

  #30 opened 6 years ago by wq2012
  1

• Have you train the model on one database and test it on another database?

  #24 opened 6 years ago by wuqiangch
  3

• Question on time cost during each iteration

  #27 opened 6 years ago by hcfeng201
  6

• about the version of pytorch and tensorflow

  #36 opened 6 years ago by Aurora11111
  1

• [Invalid][Cloud] Speaker tag is not accurate

  #34 opened 6 years ago by balavenkatesh3322
  1

• Allow `predict()` to accept a list of test_sequences as input

  #35 opened 6 years ago by wq2012
  0

• Test modell

  #31 opened 6 years ago by monosakkarid
  1

• Consider allowing enforcing max number of speakers in predict()

  #26 opened 6 years ago by wq2012
  0

• Large datasets cause training machine to run out of memory

  #8 opened 6 years ago by xinli94
  4

• wrong diarization results for long sentence (about 60 seconds)

  #25 opened 6 years ago by xebro
  1

• Publish the library as a package on PyPI

  #17 opened 6 years ago by wq2012
  0

• Better fit() API, to accept list of sequences

  #22 opened 6 years ago by wq2012
  0

• Add a `partial_fit()` API

  #19 opened 6 years ago by wq2012
  0

• Obscurity involved in sampling rate information of datasets used

  #23 opened 6 years ago by ronva-h
  1

• Dataset

  #21 opened 6 years ago by DenisSouth
  1

• Model predicts new cluster for each input after calling load()

  #13 opened 6 years ago by dalonlobo
  15

• run_test.sh problem

  #14 opened 6 years ago by 77281900000
  11

• Allow controlling verbosity level

  #16 opened 6 years ago by wq2012
  1

• How can i train my own data on this ?

  #15 opened 6 years ago by NightFury10497
  1

• Interpreting the loss values

  #12 opened 6 years ago by dalonlobo
  1

• 2 space indentation

  #11 opened 6 years ago by dalonlobo
  1

• Question on training loss

  #10 opened 6 years ago by 77281900000
  2

• Refactor fit() and predict()

  #7 opened 6 years ago by wq2012
  2

• Information about the Data

  #1 opened 6 years ago by dsleo
  3

• How to embedding audio stream data to k-vector (512)

  #9 opened 6 years ago by buaapengbo
  1

• ValueError: not enough values to unpack (expected 2, got 1)

  #6 opened 6 years ago by MuruganR96
  5

• Support half-life for learning rate

  #5 opened 6 years ago by wq2012
  0

• Pretrained Models

  #2 opened 6 years ago by dsleo
  1

• Use a real sequence accuracy evaluation instead of an approximate accuracy

  #3 opened 6 years ago by wq2012
  0