corentinj / librispeech-alignments Goto Github PK

Can u share a tool to convert alignments from the below raw files created with mfa?

File type = "ooTextFile"
Object class = "TextGrid"

xmin = 0.0
xmax = 11.89775
tiers?
size = 2
item []:
item [1]:
class = "IntervalTier"
name = "words"
xmin = 0.0
xmax = 11.89775
intervals: size = 26
intervals [1]:
xmin = 0.0
xmax = 0.120
text = ""
intervals [2]:
xmin = 0.120
xmax = 1.080
text = "인사를"
intervals [3]:
xmin = 1.080
xmax = 1.850
text = "결정하는"
intervals [4]:
xmin = 1.850
xmax = 2.650
text = "과정에서

I wanted to try and replicate these alignments, but it looks like the timestamps were different than yours. Were you using the default configuration or did you make any changes?

Thanks!

dear author，

Thanks for your code, it is convenient for processing dataset. Could I ask what are necessary requirements of wav and txt files when I process other multi-language dataset downloaded from website?

I will be appreciated if you can reply.
Good luck and Sincere blessings！

I am trying to make a dataset out of a small part of the LibriSpeech dataset to fine tune the Real-Time-Voice-Synthesis synthesizer for a single speaker. However, in order to fix an error I am trying data augmentation, adding a file to improve a particularly problematic word. When I try to call montrealforcedaligner I get

Traceback (most recent call last):
File "aligner/command_line/align.py", line 186, in
File "aligner/command_line/align.py", line 142, in validate_args
File "aligner/command_line/align.py", line 85, in align_corpus
File "aligner/corpus.py", line 543, in speaker_utterance_info
ZeroDivisionError: division by zero.

All of the files were converted to wav before this operation btw.

At least one of the files in 68MB file at https://drive.google.com/file/d/1WYfgr31T-PPwMcxuAq09XZfHQO5Mw8fE/view?usp=sharing is truncated in the middle of the list of intervals with several lines missing and no closing "

Hi,
Did you keep the phoneme alignments somewhere? It must have been computed during the process of word alignment.
Thank you :)