text/japanese_bert.py

import torch
from transformers import AutoTokenizer, AutoModelForMaskedLM
import sys
import os
from text.japanese import text2sep_kata
tokenizer = AutoTokenizer.from_pretrained("./bert/bert-base-japanese-v3")

models = dict()


def get_bert_feature(text, word2ph, device=None):
    sep_text,_ = text2sep_kata(text)
    sep_tokens = [tokenizer.tokenize(t) for t in sep_text]
    sep_ids = [tokenizer.convert_tokens_to_ids(t) for t in sep_tokens]
    sep_ids = [2]+[item for sublist in sep_ids for item in sublist]+[3]
    return get_bert_feature_with_token(sep_ids, word2ph, device)


# def get_bert_feature(text, word2ph, device=None):
#     if (
#         sys.platform == "darwin"
#         and torch.backends.mps.is_available()
#         and device == "cpu"
#     ):
#         device = "mps"
#     if not device:
#         device = "cuda"
#     if device not in models.keys():
#         models[device] = AutoModelForMaskedLM.from_pretrained(
#             "cl-tohoku/bert-base-japanese-v3"
#         ).to(device)
#     with torch.no_grad():
#         inputs = tokenizer(text, return_tensors="pt")
#         for i in inputs:
#             inputs[i] = inputs[i].to(device)
#         res = models[device](**inputs, output_hidden_states=True)
#         res = torch.cat(res["hidden_states"][-3:-2], -1)[0].cpu()
#     assert inputs["input_ids"].shape[-1] == len(word2ph)
#     word2phone = word2ph
#     phone_level_feature = []
#     for i in range(len(word2phone)):
#         repeat_feature = res[i].repeat(word2phone[i], 1)
#         phone_level_feature.append(repeat_feature)

#     phone_level_feature = torch.cat(phone_level_feature, dim=0)

#     return phone_level_feature.T

def get_bert_feature_with_token(tokens, word2ph, device=None):
    if (
        sys.platform == "darwin"
        and torch.backends.mps.is_available()
        and device == "cpu"
    ):
        device = "mps"
    if not device:
        device = "cuda"
    if device not in models.keys():
        models[device] = AutoModelForMaskedLM.from_pretrained(
            "./bert/bert-base-japanese-v3"
        ).to(device)
    with torch.no_grad():
        inputs = torch.tensor(tokens).to(device).unsqueeze(0)
        token_type_ids = torch.zeros_like(inputs).to(device)
        attention_mask = torch.ones_like(inputs).to(device)
        inputs = {"input_ids": inputs, "token_type_ids": token_type_ids, "attention_mask": attention_mask}


        # for i in inputs:
        #     inputs[i] = inputs[i].to(device)
        res = models[device](**inputs, output_hidden_states=True)
        res = torch.cat(res["hidden_states"][-3:-2], -1)[0].cpu()
    assert inputs["input_ids"].shape[-1] == len(word2ph)
    word2phone = word2ph
    phone_level_feature = []
    for i in range(len(word2phone)):
        repeat_feature = res[i].repeat(word2phone[i], 1)
        phone_level_feature.append(repeat_feature)

    phone_level_feature = torch.cat(phone_level_feature, dim=0)

    return phone_level_feature.T


if __name__ == "__main__":
    print(get_bert_feature("観覧車",[4,2]))
    pass