README.md

metadata

license: mit
pipeline_tag: text-classification
datasets:
  - thewh1teagle/phonikud-phonemes-data
  - neurlang/phonetic
language:
  - he

Goruut Extra Models

Hebrew3

PYTHON - Instructions

1. Download the model: wget https://huggingface.co/neurlang/goruut_extra_models/resolve/main/hebrew3.zip -O /tmp/hebrew3.zip
2. Run it in pygoruut pip version >= 0.6.5:

from pygoruut.pygoruut import Pygoruut

pygoruut = Pygoruut(models={"Hebrew3": "/tmp/hebrew3.zip"})

print(str(pygoruut.phonemize(language="Hebrew3", sentence="וכלה באיך לעשות שיווק נכון פרט לעצות ולתובנות שלי.")))

GOLANG - Instructions

1. Clone https://github.com/neurlang/goruut/ master branch
2. Compile goruut: cd cmd/goruut ; go build
3. Clone this repo (copy model zip to /tmp/)
4. Create a config file for goruut:

{
  "Port": "18080",
  "AdminPort": "28080",
  "PolicyMaxWords": 999999999,
  "Logging": {"Level": "debug"},
  "LoadModels": [
    {"Lang": "Hebrew", "File": "/tmp/hebrew3.zip"}
  ]
}

5. Run goruut: ./goruut --configfile my_config.json
6. Should see:

INFO[0000] Loaded language Hebrew as /tmp/hebrew3.zip   

7. Go to http://localhost:18080/
8. Infer using language Hebrew (the cloud modul is now in use)
9. Should see:

INFO[0002] Loader used file: Hebrew /tmp/hebrew3.zip language.json 
INFO[0002] Loader used file: Hebrew /tmp/hebrew3.zip language.json 
INFO[0002] Loader used file: Hebrew /tmp/hebrew3.zip language.json 
INFO[0002] Loader used file: Hebrew /tmp/hebrew3.zip weights4.json.zlib 
INFO[0002] Loader used file: Hebrew /tmp/hebrew3.zip missing.all.zlib 
INFO[0003] Loader used file: Hebrew /tmp/hebrew3.zip weights5.json.zlib