Confusing Errors
Trying to train the model got these errors:
PS C:\Users\vasil\Desktop\Stress1\beatrice-trainer> .venv\Scripts\activate
(beatrice-trainer) PS C:\Users\vasil\Desktop\Stress1\beatrice-trainer> python3 .\beatrice_trainer__main__.py -d "C:\Users\vasil\Desktop\training_data_dir" -o "C:\Users\vasil\Desktop\Output"
C:\Users\vasil\AppData\Local\Programs\Python\Python311\python.exe: can't open file 'C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py': [Errno 2] No such file or directory
(beatrice-trainer) PS C:\Users\vasil\Desktop\Stress1\beatrice-trainer> python3 beatrice_trainer_main.py -d "C:\Users\vasil\Desktop\training_data_dir" -o "C:\Users\vasil\Desktop\Output"
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer_main.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
device=cuda
config:
{'adam_betas': [0.8, 0.99],
'adam_eps': 1e-06,
'augmentation_formant_shift_probability': 0.5,
'augmentation_formant_shift_semitone_max': 3.0,
'augmentation_formant_shift_semitone_min': -3.0,
'augmentation_lpf_cutoff_freq_candidates': [2000.0, 3000.0, 4000.0, 6000.0],
'augmentation_lpf_probability': 0.2,
'augmentation_reverb_probability': 0.5,
'augmentation_snr_candidates': [20.0, 25.0, 30.0, 35.0, 40.0, 45.0],
'batch_size': 8,
'compile_convnext': False,
'compile_d4c': False,
'compile_discriminator': False,
'evaluation_interval': 2000,
'floor_noise_level': 0.001,
'grad_balancer_ema_decay': 0.995,
'grad_weight_adv': 150.0,
'grad_weight_ap': 100.0,
'grad_weight_fm': 150.0,
'grad_weight_loudness': 1.0,
'grad_weight_mel': 50.0,
'hidden_channels': 256,
'in_ir_wav_dir': 'assets/ir',
'in_noise_wav_dir': 'assets/noise',
'in_sample_rate': 16000,
'in_test_wav_dir': 'assets/test',
'learning_rate_d': 5e-05,
'learning_rate_decay': 0.999999,
'learning_rate_g': 5e-05,
'n_steps': 10000,
'num_workers': 16,
'out_sample_rate': 24000,
'phone_extractor_file': 'assets/pretrained/122_checkpoint_03000000.pt',
'phone_noise_ratio': 0.5,
'pitch_bins': 448,
'pitch_estimator_file': 'assets/pretrained/104_3_checkpoint_00300000.pt',
'pretrained_file': 'assets/pretrained/151_checkpoint_libritts_r_200_02750000.pt.gz',
'profile': False,
'record_metrics': False,
'san': False,
'save_interval': 2000,
'segment_length': 100,
'training_time_vq': 'none',
'use_amp': True,
'vq_topk': 4,
'warmup_steps': 5000,
'wav_length': 96000}
n_speakers=1
0: Haru
len(training_filelist)=286
len(test_filelist)=8
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\assets\test\common_voice_ja_38833628_16k.wav, [0]
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\assets\test\common_voice_ja_38843402_16k.wav, [0]
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\assets\test\common_voice_ja_38852485_16k.wav, [0]
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\assets\test\common_voice_ja_38853932_16k.wav, [0]
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\assets\test\common_voice_ja_38864552_16k.wav, [0]
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\assets\test\common_voice_ja_38878413_16k.wav, [0]
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\assets\test\common_voice_ja_38898180_16k.wav, [0]
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\assets\test\common_voice_ja_38925334_16k.wav, [0]
C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torchaudio_backend\utils.py:213: UserWarning: In 2.9, this function's implementation will be changed to use torchaudio.load_with_torchcodecunder the hood. Some parameters like ``normalize``, ``format``, ``buffer_size``, and ``backend`` will be ignored. We recommend that you port your code to rely directly on TorchCodec's decoder instead: https://docs.pytorch.org/torchcodec/stable/generated/torchcodec.decoders.AudioDecoder.html#torchcodec.decoders.AudioDecoder. warnings.warn( Computing mean F0s of target speakers... 0: 129.6Hz, Done. Computing pitch shifts for test files... Computing pitch shifts: 100%|████████████████████████████████████████████████████████████| 8/8 [00:08<00:00, 1.10s/it] Done. C:\Users\vasil\Desktop\Stress1\beatrice-trainer\.venv\Lib\site-packages\torch\nn\utils\weight_norm.py:144: FutureWarning:torch.nn.utils.weight_normis deprecated in favor oftorch.nn.utils.parametrizations.weight_norm. WeightNorm.apply(module, name, dim) <All keys matched successfully> <All keys matched successfully> <All keys matched successfully> <All keys matched successfully> Building codebooks: 0%| | 0/1 [00:00<?, ?it/s]C:\Users\vasil\Desktop\Stress1\beatrice-trainer\.venv\Lib\site-packages\torchaudio\_backend\utils.py:213: UserWarning: In 2.9, this function's implementation will be changed to use torchaudio.load_with_torchcodec under the hood. Some parameters like normalize, format, buffer_size, and backend will be ignored. We recommend that you port your code to rely directly on TorchCodec's decoder instead: https://docs.pytorch.org/torchcodec/stable/generated/torchcodec.decoders.AudioDecoder.html#torchcodec.decoders.AudioDecoder.
warnings.warn(
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer_main_.py:835: UserWarning: wav_length % 160 != 0
warnings.warn("wav_length % 160 != 0")
Building codebooks: 100%|████████████████████████████████████████████████████████████████| 1/1 [00:21<00:00, 21.11s/it]
Using cache found in C:\Users\vasil/.cache\torch\hub\tarepan_SpeechMOS_v1.0.0
C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\nn\utils\weight_norm.py:144: FutureWarning: torch.nn.utils.weight_norm is deprecated in favor of torch.nn.utils.parametrizations.weight_norm.
WeightNorm.apply(module, name, dim)
Training: 0%| | 0/10000 [00:00<?, ?it/s]C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer__main__.py:33: UserWarning: torchaudio._backend.list_audio_backends has been deprecated. This deprecation is part of a large refactoring effort to transition TorchAudio into a maintenance phase. The decoding and encoding capabilities of PyTorch for both audio and video are being consolidated into TorchCodec. Please see https://github.com/pytorch/audio/issues/3902 for more information. It will be removed from the 2.9 release.
assert "soundfile" in torchaudio.list_audio_backends()
Traceback (most recent call last):
File "", line 1, in
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\multiprocessing\spawn.py", line 122, in spawn_main
exitcode = _main(fd, parent_sentinel)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\multiprocessing\spawn.py", line 131, in _main
prepare(preparation_data)
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\multiprocessing\spawn.py", line 246, in prepare
_fixup_main_from_path(data['init_main_from_path'])
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\multiprocessing\spawn.py", line 297, in fixup_main_from_path
main_content = runpy.run_path(main_path,
^^^^^^^^^^^^^^^^^^^^^^^^^
File "", line 291, in run_path
File "", line 98, in run_module_code
File "", line 88, in run_code
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer\beatrice_trainer_main.py", line 25, in
import torch
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch_init.py", line 2150, in
from torch import VF as VF, functional as functional # usort: skip
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\functional.py", line 8, in
import torch.nn.functional as F
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\nn_init.py", line 8, in
from torch.nn.modules import * # usort: skip # noqa: F403
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\nn\modules_init.py", line 35, in
from .batchnorm import (
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\nn\modules\batchnorm.py", line 9, in
from .functions import SyncBatchNorm as sync_batch_norm
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\nn\modules_functions.py", line 4, in
from torch.autograd.function import Function
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\autograd_init.py", line 32, in
from .gradcheck import gradcheck, gradgradcheck
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\autograd\gradcheck.py", line 11, in
import torch.testing
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\testing_init.py", line 4, in
from ._comparison import assert_allclose, assert_close as assert_close
File "C:\Users\vasil\Desktop\Stress1\beatrice-trainer.venv\Lib\site-packages\torch\testing_comparison.py", line 168, in
abs_diff_idx: Optional[Union[int, tuple[int, ...]]] = None,
~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 376, in inner
return cached(*args, **kwds)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 502, in getitem
return self._getitem(self, parameters)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 727, in Optional
return Union[arg, type(None)]
~~~~~^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 376, in inner
return cached(*args, **kwds)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 502, in getitem
return self._getitem(self, parameters)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 715, in Union
parameters = tuple(_type_check(p, msg) for p in parameters)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 715, in
parameters = tuple(_type_check(p, msg) for p in parameters)
^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 186, in _type_check
arg = _type_convert(arg, module=module, allow_special_forms=allow_special_forms)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\vasil\AppData\Local\Programs\Python\Python311\Lib\typing.py", line 159, in _type_convert
def _type_convert(arg, module=None, *, allow_special_forms=False):
KeyboardInterrupt