blackteay/GPT-SoVITS
1
0
Fork 0
mirror of https://github.com/RVC-Boss/GPT-SoVITS.git synced 2025-04-25 02:58:33 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #31 from AnyaCoder/patch-4

Browse Source 
Update dataset.py
This commit is contained in:
RVC-Boss 2024-01-17 20:28:15 +08:00 committed by GitHub
commit 7e6816fff2
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 3 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
GPT_SoVITS/AR/data/dataset.py
View File

@ -116,6 +116,7 @@ class Text2SemanticDataset(Dataset):
phoneme_data_len = len(self.phoneme_data.keys()) phoneme_data_len = len(self.phoneme_data.keys())
print("semantic_data_len:", semantic_data_len) print("semantic_data_len:", semantic_data_len)
print("phoneme_data_len:", phoneme_data_len) print("phoneme_data_len:", phoneme_data_len)
print(self.semantic_data)
idx = 0 idx = 0
num_not_in = 0 num_not_in = 0
num_deleted_bigger = 0 num_deleted_bigger = 0
@ -123,7 +124,7 @@ class Text2SemanticDataset(Dataset):
for i in range(semantic_data_len): for i in range(semantic_data_len):
# 先依次遍历 # 先依次遍历
# get str # get str
item_name = self.semantic_data["item_name"][i] item_name = self.semantic_data.iloc[i,0]
# print(self.phoneme_data) # print(self.phoneme_data)
try: try:
phoneme, word2ph, text = self.phoneme_data[item_name] phoneme, word2ph, text = self.phoneme_data[item_name]
@ -133,7 +134,7 @@ class Text2SemanticDataset(Dataset):
num_not_in += 1 num_not_in += 1
continue continue
semantic_str = self.semantic_data["semantic_audio"][i] semantic_str = self.semantic_data.iloc[i,1]
# get token list # get token list
semantic_ids = [int(idx) for idx in semantic_str.split(" ")] semantic_ids = [int(idx) for idx in semantic_str.split(" ")]
# (T), 是否需要变成 (1, T) -> 不需要，因为需要求 len # (T), 是否需要变成 (1, T) -> 不需要，因为需要求 len