microsoft/UniSpeech

Issues

• error when pip install and cannot import name 'Wav2VecModel' from 'fairseq.models.wav2vec

  #26 opened 3 years ago by leijue222
  9

• Weighted Sum for Downstream Tasks

  #58 opened 3 months ago by jaeyeonkim99
  0

• diarization - KeyError: 'embed.weight'

  #21 opened 3 years ago by RuslanSel
  5

• Model checkpoints not available

  #56 opened 9 months ago by xulinfan
  1

• Speaker verification result

  #46 opened 2 years ago by pierfale
  4

• error when torch.hub.load('s3prl/s3prl:e52439edaeb1a443e82960e6401ae6ab4241def6', "wavlm_large")

  #48 opened 2 years ago by jiazj-jiazj
  3

• [Downstream Tasks] - Diarization using WavLM

  #57 opened 9 months ago by ai-nikolai
  0

• Time stamps in out.rttm file for speaker_diarization

  #30 opened 3 years ago by bryant0918
  1

• Missing pretrained model for ecapa-tdnn

  #55 opened 9 months ago by MichelMarieLamah13
  0

• Problem of cosine similarity computing for speaker verification

  #47 opened 2 years ago by Winterspringkle
  1

• Inquiry about Continual Training with WavLM and Pre-Training Resources

  #54 opened a year ago by CantaoSu
  0

• pip install --require-hashes -r requirements.txt

  #52 opened a year ago by evelynyhc
  1

• The link for the pretrained model is unavailable #UniSpeech-SAT Large

  #50 opened a year ago by TKsavy
  3

• Why is my duplicated wavLM results on vox1-o is 30% worse

  #28 opened 3 years ago by AIDman
  3

• training wavlm

  #49 opened a year ago by dutchsing009
  0

• How to use finetune wavlm large model?

  #40 opened 2 years ago by haha010508
  2

• This repo is missing important files

  #44 opened 2 years ago by microsoft-github-policy-service
  0

• source code to train speaker verification with WavLM-large pre-trained model

  #42 opened 2 years ago by dntuong
  0

• About finetuning HuBERT_base_ls_960.pt with 1h data

  #41 opened 2 years ago by LYPinASR
  0

• wavlm model for downstream test?

  #38 opened 2 years ago by kli017
  0

• source code to train speaker diarization with the pre-trained feature extract model?

  #35 opened 2 years ago by kli017
  0

• UniSpeech Model Download

  #37 opened 2 years ago by hongfeixue
  0

• pre-training detail for Unispeech-SAT

  #36 opened 3 years ago by Sreyan88
  0

• Is "unispeech_sat.th" wrong ?

  #14 opened 3 years ago by Damien-Da
  15

• Hi, how to calculate the EER or DER

  #34 opened 3 years ago by liyunlongaaa
  0

• How continue pretrain WavLM for ASR?

  #33 opened 3 years ago by cnlinxi
  0

• Hello, I want to do the voiceprint verification task, the model chooses unispeech_sat, but it keeps getting an error, please help

  #31 opened 3 years ago by zuihaodeyujian
  5

• about updated file "unispeech_sat.th"

  #32 opened 3 years ago by zuihaodeyujian
  0

• WavLM Inference Error

  #29 opened 3 years ago by bryant0918
  2

• The code for computing the relative positional embedding

  #25 opened 3 years ago by mycrazycracy
  1

• Where is the trainning/test/infer recipe?

  #27 opened 3 years ago by leijue222
  1

• Bug in WavLM?

  #24 opened 3 years ago by OlaWod
  1

• Pre-processed data missing

  #23 opened 3 years ago by umbertomichieli
  0

• KeyError: "'_name'" Speaker Diarization

  #22 opened 3 years ago by taimoorkhokhar
  1

• Unispeech-SAT fairseq code

  #20 opened 3 years ago by RuslanSel
  4

• Formula 6 in paper

  #16 opened 3 years ago by Sreyan88
  7

• Missing recipe/code for speaker verification

  #18 opened 3 years ago by ajinkyakulkarni14
  1

• UniSpeech-SAT/speaker_verification: torchaudio version

  #17 opened 3 years ago by msh9184
  1

• Huggingface sat model missing tokenizer

  #15 opened 3 years ago by bagustris
  4

• Getting speaker embeddings

  #13 opened 3 years ago by AH289
  3

• Fine-tuned Unispeech models cannot be downloaded

  #9 opened 3 years ago by patrickvonplaten
  1

• Speech transcription example?

  #7 opened 3 years ago by johnpaulbin
  1

• Release vocab.json for Common Voice

  #3 opened 3 years ago by patrickvonplaten
  3

• More details about the output

  #6 opened 3 years ago by AhmedHashish123
  2

• Access Required for UniSpeech-SAT models

  #4 opened 3 years ago by patrickvonplaten
  2