mesolitica/malaya-speech

Issues

• Link to wikipedia dataset seems to be the same as IIUM.

  #45 opened 4 months ago by kwatcharasupat
  1

• NOT euqal to the article

  #2 opened 3 years ago by hdmjdp
  3

• NER, long audio, multilingual

  #44 opened a year ago by hududed
  0

• Teme audiobook

  #40 opened a year ago by huseinzol05
  1

• BUKAN KERANA AKU TAK CINTA audiobook

  #39 opened a year ago by huseinzol05
  1

• harry potter audiobook

  #38 opened a year ago by huseinzol05
  1

• trained speaker count detector

  #43 opened a year ago by huseinzol05
  0

• Prepare Zipformer model scripts

  #36 opened a year ago by huseinzol05
  1

• remove librosa as dependency

  #42 opened a year ago by huseinzol05
  0

• calculate total hours parsed dataset

  #34 opened a year ago by huseinzol05
  0

• calculate total hours postfilter parsed dataset

  #37 opened a year ago by huseinzol05
  0

• parse dataset

  #33 opened a year ago by huseinzol05
  0

• Can you share you bestrq training log?

  #27 opened a year ago by Mddct
  0

• download from youtube

  #41 opened a year ago by huseinzol05
  0

• postfilter parsed dataset

  #35 opened a year ago by huseinzol05
  0

• calculate logprob distribution plot parsed dataset

  #32 opened a year ago by huseinzol05
  0

• semisupervised youtube videos using Whisper Large V2

  #31 opened a year ago by huseinzol05
  0

• need to fix speaking

  #30 opened a year ago by vqoley
  2

• Problem in the Pipeline class

  #29 opened a year ago by guilhermebernieri
  1

• Error while loading the model

  #28 opened a year ago by gowthambg20
  1

• There is no spectral_cluster

  #26 opened 2 years ago by codeHorasan
  1

• Need some help in convert-asr-to-tflite

  #25 opened 2 years ago by Ajay031998
  1

• Ask for tips in conformer model training

  #24 opened 2 years ago by mr-coconut
  1

• Migration Large-Singlish-Conformer pretrained checkpoint from TF1 to TF2

  #22 opened 2 years ago by mr-coconut
  1

• Dataset Problem

  #23 opened 2 years ago by umerghayaz
  1

• Retraining from output-large-singlish-conformer checkpoint

  #21 opened 2 years ago by mr-coconut
  4

• Retraining of Singlish-Large-Conformer

  #20 opened 2 years ago by mr-coconut
  1

• How does the Semisupervised malay dataset do vad?

  #19 opened 2 years ago by 283258771
  1

• Each run of greedy_decode may generate different result

  #18 opened 2 years ago by Beast-Leon
  7

• Noise_reduction.deep_model

  #17 opened 2 years ago by guilhermebernieri
  2

• any example code about the how to do the quantize?

  #15 opened 2 years ago by tricky61
  1

• No file named 'base_directory/haqkiem/metadata.csv'

  #16 opened 2 years ago by EdwinKuo1337
  1

• Copy BackBlaze to Hugging Face

  #14 opened 2 years ago by huseinzol05
  1

• Can update SD module to adapt to "spectralcluster" latest version?

  #13 opened 3 years ago by ZhaoTTT
  4

• Can we have the tf.saved_model pb models (tensorflow 2)?

  #12 opened 3 years ago by rizwanishaq
  1

• Error while defining model

  #10 opened 3 years ago by Anuj040
  2

• Issue with large-conformer

  #11 opened 3 years ago by Anuj040
  0

• Have installed malaya but still having the ImportError

  #9 opened 3 years ago by MaisarahPauzi
  1

• Problems when loading model

  #8 opened 3 years ago by hzqh1
  2

• The link of Universal MelGAN tar.gz seems to be missed, could you so kind to provide it? XD

  #6 opened 3 years ago by Liofang
  2

• Error When Load Vocoder Model..

  #5 opened 3 years ago by FsalKml
  3

• Quantized models raises an error

  #4 opened 3 years ago by hmtkvs
  1

• how's the rtf of 1024 filters on cpu?

  #3 opened 3 years ago by hdmjdp
  1

• Support TF 2

  #1 opened 4 years ago by huseinzol05
  1