Edit on GitHub
  • John Snow Labs
  • Oct 19, 2023

Uzbek asr_whisper_small_uzbek_pipeline pipeline WhisperForCTC from BlueRaccoon

  • whisper
  • uz
  • open_source
  • pipeline

Description

Pretrained WhisperForCTC, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP.asr_whisper_small_uzbek_pipeline is a Uzbek model originally trained by BlueRaccoon.

This model is only compatible with PySpark 3.4 and above

Download Copy S3 URI

How to use


pipeline = PretrainedPipeline('asr_whisper_small_uzbek_pipeline', lang = 'uz')
annotations =  pipeline.transform(audioDF)



val pipeline = new PretrainedPipeline('asr_whisper_small_uzbek_pipeline', lang = 'uz')
val annotations = pipeline.transform(audioDF)

Model Information

Model Name: asr_whisper_small_uzbek_pipeline
Type: pipeline
Compatibility: Spark NLP 5.1.4+
License: Open Source
Edition: Official
Language: uz
Size: 1.7 GB

References

https://huggingface.co/BlueRaccoon/whisper-small-uz

Included Models

  • AudioAssembler
  • WhisperForCTC