Description
Pretrained Wav2vec2 pipeline, adapted from Hugging Face and curated to provide scalability and production-readiness using Spark NLP.asr_wav2vec2_large_xlsr_53_dutch_by_facebook is a Dutch, Flemish model originally trained by facebook.
NOTE: This pipeline only works on a CPU, if you need to use this pipeline on a GPU device please use pipeline_asr_wav2vec2_large_xlsr_53_dutch_by_facebook
How to use
pipeline = PretrainedPipeline('pipeline_asr_wav2vec2_large_xlsr_53_dutch_by_facebook_gpu', lang = 'nl')
annotations = pipeline.transform(audioDF)
val pipeline = new PretrainedPipeline("pipeline_asr_wav2vec2_large_xlsr_53_dutch_by_facebook_gpu", lang = "nl")
val annotations = pipeline.transform(audioDF)
Model Information
| Model Name: | pipeline_asr_wav2vec2_large_xlsr_53_dutch_by_facebook_gpu |
| Type: | pipeline |
| Compatibility: | Spark NLP 4.2.0+ |
| License: | Open Source |
| Edition: | Official |
| Language: | nl |
| Size: | 756.4 MB |
Included Models
- AudioAssembler
- Wav2Vec2ForCTC