Edit on GitHub
  • John Snow Labs
  • Apr 01, 2023

Korean asr_swin_distilhubert_ko_zeroth TFHubertForCTC from Bingsu

  • hubert
  • ko
  • open_source
  • audio
  • asr
  • pipeline

Description

Pretrained Hubert pipeline, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP.asr_swin_distilhubert_ko_zeroth is a Korean model originally trained by Bingsu.

Download Copy S3 URI

How to use


    pipeline = PretrainedPipeline('pipeline_asr_swin_distilhubert_ko_zeroth', lang = 'ko')
    annotations =  pipeline.transform(audioDF)
    


    val pipeline = new PretrainedPipeline("pipeline_asr_swin_distilhubert_ko_zeroth", lang = "ko")
    val annotations = pipeline.transform(audioDF)

Model Information

Model Name: pipeline_asr_swin_distilhubert_ko_zeroth
Type: pipeline
Compatibility: Spark NLP 4.4.0+
License: Open Source
Edition: Official
Language: ko
Size: 183.6 MB

Included Models

  • AudioAssembler
  • HubertForCTC