Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing

  • Log In
  • Sign Up

marcoyang
/
spear-encoder-streaming-600M-speech-only

Model card Files Files and versions
xet
Community
spear-encoder-streaming-600M-speech-only
7.19 GB
Ctrl+K
Ctrl+K
  • 1 contributor
History: 9 commits
marcoyang's picture
marcoyang
add real chunk-wise streaming inference code
0bdcf6a 6 months ago
  • spear-xlarge-non-streaming
    support inference non-streaming models 6 months ago
  • v0.1
    add note 6 months ago
  • v0.2
    add note 6 months ago
  • .gitattributes
    1.52 kB
    initial commit 7 months ago
  • 1284-1180-0027.flac
    72.5 kB
    init 7 months ago
  • README.md
    31 Bytes
    initial commit 7 months ago
  • envirnoment.txt
    25 Bytes
    init 7 months ago
  • inference_600m.py
    8.73 kB
    remove lhotse dependency, use torchaudio for fbank computation 6 months ago
  • inference_600m.sh
    390 Bytes
    support inference non-streaming models 6 months ago
  • inference_600m_non_streaming.sh
    252 Bytes
    support inference non-streaming models 6 months ago
  • inference_600m_streaming_forward.py
    20.6 kB
    add real chunk-wise streaming inference code 6 months ago
  • inference_600m_streaming_forward.sh
    363 Bytes
    add real chunk-wise streaming inference code 6 months ago
  • model.py
    31.4 kB
    support returning layer-wise hidden states 6 months ago
  • scaling.py
    71 kB
    init 7 months ago
  • subsampling.py
    13.1 kB
    init 7 months ago
  • utilities.py
    2.89 kB
    init 7 months ago
  • zipformer.py
    95.7 kB
    add real chunk-wise streaming inference code 6 months ago