MontrealCorpusTools
[BUG] FeatureGenerationError on soundfiles
Debugging checklist
[x] Have you updated to latest MFA version? Yes
[x] Have you tried rerunning the command with the --clean flag? Yes
Describe the issue
I want to generate TextGrids from my own soundfiles (I don't have textfiles for them). No problem when using the mfa validate command.
For Reproducing your issue Please fill out the following:
- Corpus structure
- What language is the corpus in? French
- How many files/speakers? 18/3
- Are you using lab files or TextGrid files for input? None
- Dictionary
- Are you using a dictionary from MFA? If so, which one? french_mfa
- If it's a custom dictionary, what is the phoneset?
- Acoustic model
- If you're using an acoustic model, is it one download through MFA? If so, which one? french_mfa
- If it's a model you've trained, what data was it trained on?
Log file
INFO Generating final features...
0% ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0/18 [ 0:00:01 < -:--:-- , ? it/s ]
ERROR There was an error in the run, please see the log.
Exception ignored in atexit callback: <bound method ExitHooks.history_save_handler of <montreal_forced_aligner.command_line.mfa.ExitHooks object at 0x7f984d21a910>>
Traceback (most recent call last):
File "/home/cbeaumard/miniconda3/envs/aligner/lib/python3.11/site-packages/montreal_forced_aligner/command_line/mfa.py", line 100, in history_save_handler
raise self.exception
File "/home/cbeaumard/miniconda3/envs/aligner/bin/mfa", line 10, in
No utterances had features, please check the logs for errors.
Desktop (please complete the following information):
- OS: Windows
- Version: Windows 11
- Any other details about the setup (Cloud, Docker, etc): server connected with SSH (under linux)
Additional information MFA was installed with conda.
Also have the same error
- System: Intel and M2 Mac
- MFA latest (running fine on other files); installed using conda and in the same environment as openai-whisper
- WAV files in English
- english_mfa acoustic model and english_us_mfa dictionary
- validation workflow not tested for this bug
Are there any errors listed in the make_mfcc.X.log, generate_final_features.X.log, or cmvn.log files in ~/Documents/MFA/{corpus_name}/{corpus_name}/splitX/log (replacing {corpus_name} with your dataset name and X with the number of jobs)?
This is likely a bug in 3.0.X only, so if you need a workable version asap, installing the stable 2.2 via https://montreal-forced-aligner.readthedocs.io/en/latest/installation.html#installing-older-versions-of-mfa should unblock you.
@mmcauliffe Can confirm that this is solved when moving to 2.2.17, so most likely a bug in 3.0.X. Thank you for the quick response.
I also meet this bug, version is 3.0.X alpha 6:
mfa align /scratch/v-zhikangniu/LibriSpeech english_us_mfa english_mfa /data/v-zhikangniu/algner --clean
Please be aware that you are running an alpha version of MFA. If you would like to install a more stable version, please visit https://montreal-forced-aligner.readthedocs.io/en/latest/installation.html#installing-older-versions-of-mfa
INFO Setting up corpus information...
INFO Loading corpus from source files...
100% ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 433,343/100 [ 0:00:00 < 0:00:00 , 2,383 it/s ]
INFO Found 5831 speakers across 433343 files, average number of utterances per speaker: 74.31709826787858
INFO Initializing multiprocessing jobs...
INFO Normalizing text...
0% ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0/433,343 [ 0:00:02 < -:--:-- , ? it/s ]
INFO Generating MFCCs...
0% ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0/433,343 [ 0:00:02 < -:--:-- , ? it/s ]
INFO Calculating CMVN...
INFO Generating final features...
0% ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0/433,343 [ 0:00:02 < -:--:-- , ? it/s ]
ERROR There was an error in the run, please see the log.
Exception ignored in atexit callback: <bound method ExitHooks.history_save_handler of <montreal_forced_aligner.command_line.mfa.ExitHooks object at 0x7fbf16508dd0>>
Traceback (most recent call last):
File "/anaconda/envs/aligner/lib/python3.11/site-packages/montreal_forced_aligner/command_line/mfa.py", line 107, in history_save_handler
raise self.exception
File "/anaconda/envs/aligner/bin/mfa", line 10, in <module>
sys.exit(mfa_cli())
^^^^^^^^^
File "/anaconda/envs/aligner/lib/python3.11/site-packages/click/core.py", line 1157, in __call__
return self.main(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/anaconda/envs/aligner/lib/python3.11/site-packages/rich_click/rich_command.py", line 126, in main
rv = self.invoke(ctx)
^^^^^^^^^^^^^^^^
File "/anaconda/envs/aligner/lib/python3.11/site-packages/click/core.py", line 1688, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/anaconda/envs/aligner/lib/python3.11/site-packages/click/core.py", line 1434, in invoke
return ctx.invoke(self.callback, **ctx.params)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/anaconda/envs/aligner/lib/python3.11/site-packages/click/core.py", line 783, in invoke
return __callback(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/anaconda/envs/aligner/lib/python3.11/site-packages/click/decorators.py", line 33, in new_func
return f(get_current_context(), *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/anaconda/envs/aligner/lib/python3.11/site-packages/montreal_forced_aligner/command_line/align.py", line 122, in align_corpus_cli
aligner.align()
File "/anaconda/envs/aligner/lib/python3.11/site-packages/montreal_forced_aligner/alignment/pretrained.py", line 333, in align
self.setup()
File "/anaconda/envs/aligner/lib/python3.11/site-packages/montreal_forced_aligner/alignment/pretrained.py", line 207, in setup
self.load_corpus()
File "/anaconda/envs/aligner/lib/python3.11/site-packages/montreal_forced_aligner/corpus/acoustic_corpus.py", line 1089, in load_corpus
self.generate_features()
File "/anaconda/envs/aligner/lib/python3.11/site-packages/montreal_forced_aligner/corpus/acoustic_corpus.py", line 511, in generate_features
self.generate_final_features()
File "/anaconda/envs/aligner/lib/python3.11/site-packages/montreal_forced_aligner/corpus/acoustic_corpus.py", line 458, in generate_final_features
raise FeatureGenerationError(
montreal_forced_aligner.exceptions.FeatureGenerationError: FeatureGenerationError:
No utterances had features, please check the logs in /home/v-zhikangniu/Documents/MFA/LibriSpeech/LibriSpeech/split3/log for errors.
and the mfa validate looks like successful?
I was facing the same issue for multi-speaker folders. when i did the alignment for individual files the error was no more. So i do not understand what is happening here.
