syhw/wer_are_we

Issues

• why don't update large model in audio

  #60 opened 4 months ago by liuanping
  0

• Why stopped updating?

  #59 opened a year ago by wwfcnu
  1

• i don't see anything new till now. how about big model in audio

  #58 opened 2 years ago by liuanping
  0

• "WER are we" for other languages than English?

  #51 opened 4 years ago by bmilde
  2

• New sota on librispeech

  #53 opened 3 years ago by yzhang87
  0

• Add `speech-recognition` tag to the repo

  #50 opened 4 years ago by abitrolly
  1

• Include "Conformer: Convolution-augmented Transformer for Speech Recognition"

  #46 opened 4 years ago by est31
  1

• Request for note next to pytorch-kaldi TIMIT results

  #47 opened 4 years ago by sdrobert
  2

• new sota on Librispeech with semi-supervised learning

  #48 opened 4 years ago by bobchennan
  0

• Add CER for mandarin dataset?

  #40 opened 5 years ago by wd929
  1

• Scaling this effort up

  #32 opened 6 years ago by syhw
  2

• Links to datasets

  #6 opened 7 years ago by guillaume-chevalier
  4

• Including Kaldi results

  #31 opened 6 years ago by syhw
  2

• Do you need a maintainer?

  #28 opened 6 years ago by zcaceres
  2

• TIMIT 13.8

  #30 opened 6 years ago by TParcollet
  1

• Wav2Letter results

  #15 opened 6 years ago by igormq
  2

• cloudwalk tech, the facial imagining company claims Pyramidal-FSMN, 2.97 percent,

  #22 opened 6 years ago by dpny518
  1

• WSJ update

  #21 opened 6 years ago by smolendawid
  0

• TIMIT update

  #20 opened 6 years ago by vanekyj
  0

• Recognizing Overlapped Speech in Meetings: A Multichannel Separation Approach Using Neural Networks

  #19 opened 6 years ago by WilliamTambellini
  0

• librispeech results

  #17 opened 6 years ago by danpovey
  1

• New dataset by Mozilla

  #7 opened 7 years ago by guillaume-chevalier
  0

• GramCTC results

  #4 opened 8 years ago by sancha
  0