Description
Pretrained WhisperForCTC, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP.asr_whisper_malayalam_first_model_pipeline
is a Malayalam model originally trained by kurianbenoy.
This model is only compatible with PySpark 3.4 and above
How to use
pipeline = PretrainedPipeline('asr_whisper_malayalam_first_model_pipeline', lang = 'ml')
annotations = pipeline.transform(audioDF)
val pipeline = new PretrainedPipeline('asr_whisper_malayalam_first_model_pipeline', lang = 'ml')
val annotations = pipeline.transform(audioDF)
Model Information
Model Name: | asr_whisper_malayalam_first_model_pipeline |
Type: | pipeline |
Compatibility: | Spark NLP 5.1.4+ |
License: | Open Source |
Edition: | Official |
Language: | ml |
Size: | 391.1 MB |
References
https://huggingface.co/kurianbenoy/whisper-ml-first-model
Included Models
- AudioAssembler
- WhisperForCTC