Lumi-Bert-VITS2 / text /chinese_bert.py
XzJosh's picture
Upload chinese_bert.py
c65ba09
import torch
from transformers import AutoTokenizer, AutoModelForMaskedLM
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
tokenizer = AutoTokenizer.from_pretrained("./bert/chinese-roberta-wwm-ext-large")
model = AutoModelForMaskedLM.from_pretrained("./bert/chinese-roberta-wwm-ext-large").to(device)
def get_bert_feature(text, word2ph):
with torch.no_grad():
inputs = tokenizer(text, return_tensors='pt')
for i in inputs:
inputs[i] = inputs[i].to(device)
res = model(**inputs, output_hidden_states=True)
res = torch.cat(res['hidden_states'][-3:-2], -1)[0].cpu()
assert len(word2ph) == len(text)+2
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__':
# feature = get_bert_feature('你好,我是说的道理。')
import torch
word_level_feature = torch.rand(38, 1024) # 12个词,每个词1024维特征
word2phone = [1, 2, 1, 2, 2, 1, 2, 2, 1, 2, 2, 1, 2, 2, 2, 2, 2, 1, 1, 2, 2, 1, 2, 2, 2, 2, 1, 2, 2, 2, 2, 2, 1, 2, 2, 2, 2, 1]
# 计算总帧数
total_frames = sum(word2phone)
print(word_level_feature.shape)
print(word2phone)
phone_level_feature = []
for i in range(len(word2phone)):
print(word_level_feature[i].shape)
# 对每个词重复word2phone[i]次
repeat_feature = word_level_feature[i].repeat(word2phone[i], 1)
phone_level_feature.append(repeat_feature)
phone_level_feature = torch.cat(phone_level_feature, dim=0)
print(phone_level_feature.shape) # torch.Size([36, 1024])