Filter unsupported languages

This commit is contained in:
starylan 2024-05-18 23:10:14 +08:00
parent 82a5672361
commit 0e08f12bed
2 changed files with 11 additions and 7 deletions

View File

@ -117,12 +117,15 @@ if os.path.exists(txt_path) == False:
try: try:
wav_name, spk_name, language, text = line.split("|") wav_name, spk_name, language, text = line.split("|")
# todo.append([name,text,"zh"]) # todo.append([name,text,"zh"])
todo.append( if language in language_v1_to_language_v2.keys():
[wav_name, text, language_v1_to_language_v2.get(language, language)] todo.append(
) [wav_name, text, language_v1_to_language_v2.get(language, language)]
)
else:
print(f"\033[33m[Waring] The {language = } of {wav_name} is not supported for training.\033[0m")
except: except:
print(line, traceback.format_exc()) print(line, traceback.format_exc())
assert len(todo) > 0, f"\033[31m[Error] Part {i_part}: No data loaded from {inp_text}.\033[0m"
process(todo, res) process(todo, res)
opt = [] opt = []
for name, phones, word2ph, norm_text in res: for name, phones, word2ph, norm_text in res:

View File

@ -411,9 +411,10 @@ def open1a(inp_text,inp_wav_dir,exp_name,gpu_numbers,bert_pretrained_dir):
opt = [] opt = []
for i_part in range(all_parts): for i_part in range(all_parts):
txt_path = "%s/2-name2text-%s.txt" % (opt_dir, i_part) txt_path = "%s/2-name2text-%s.txt" % (opt_dir, i_part)
with open(txt_path, "r", encoding="utf8") as f: if os.path.exists(txt_path):
opt += f.read().strip("\n").split("\n") with open(txt_path, "r", encoding="utf8") as f:
os.remove(txt_path) opt += f.read().strip("\n").split("\n")
os.remove(txt_path)
path_text = "%s/2-name2text.txt" % opt_dir path_text = "%s/2-name2text.txt" % opt_dir
with open(path_text, "w", encoding="utf8") as f: with open(path_text, "w", encoding="utf8") as f:
f.write("\n".join(opt) + "\n") f.write("\n".join(opt) + "\n")