doublesizebed

/

G2P_malay

like 0

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("doublesizebed/G2P_malay")
model = AutoModelForSeq2SeqLM.from_pretrained("doublesizebed/G2P_malay")

def predict(word):
inputs = tokenizer(word, return_tensors="pt")
outputs = model.generate(**inputs)
phonemes = tokenizer.decode(outputs[0], skip_special_tokens=True)
return phonemes