yaofanguk / video-subtitle-generator Goto Github PK

这个是什么问题： Failed to open file b'C:\Users\\xe5\x9d\xa4\xe5\x9d\xa4\xe5\x93\x8e\AppData\Local\Temp\scipy-6mn3p0vy'

使用您提供的配置，但是最后发现字幕生成很慢，cpu占用率很高，CUDA和显存都没有占用。如何设置以确保在GPU中运行？

我可以正常运行,可是发现是通过CPU进行计算处理的,速度比较慢,请问如何通过显卡进行处理？
显卡: 英伟达2080

GPU: Tesla P4
pyTorch: 2.0.1 (CUDA 11.8).

I first set up a new conda environment following the instructions, but the model was running on the CPU.

I then installed pyTorch CUDA using the command generated on the official website. Now running vsg throws the following error.

Exception in thread Thread-1:
Traceback (most recent call last):
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\threading.py", line 932, in _bootstrap_inner
    self.run()
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\threading.py", line 870, in run
    self._target(*self._args, **self._kwargs)
  File "gui.py", line 173, in task
    from backend import main
  File "D:\Program Files\Video Subtitle Generator\backend\main.py", line 15, in <module>
    import librosa
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\librosa\__init__.py", line 209, in <module>
    from . import core
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\librosa\core\__init__.py", line 5, in <module>
    from .convert import *  # pylint: disable=wildcard-import
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\librosa\core\convert.py", line 7, in <module>
    from . import notation
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\librosa\core\notation.py", line 8, in <module>
    from ..util.exceptions import ParameterError
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\librosa\util\__init__.py", line 77, in <module>
    from .utils import *  # pylint: disable=wildcard-import
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\librosa\util\utils.py", line 9, in <module>
    import numba
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\numba\__init__.py", line 42, in <module>
    from numba.np.ufunc import (vectorize, guvectorize, threading_layer,
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\numba\np\ufunc\__init__.py", line 3, in <module>
    from numba.np.ufunc.decorators import Vectorize, GUVectorize, vectorize, guvectorize
  File "C:\Users\USERNAME\.conda\envs\vsgEnv\lib\site-packages\numba\np\ufunc\decorators.py", line 3, in <module>
    from numba.np.ufunc import _internal
SystemError: initialization of _internal failed without raising an exception

Is there any specific pyTorch version/configuration I should use? Thanks!

改了一圈发现改不对，不知道是不是没有找到正确的位置。
How to specify a certain language instead of automatic detection?

使用 medium 模型
在日本AV片中，日文转中文过程中。
出现：谢谢大家
字幕君
求打赏
很多这样的私货。
这是什么原因？
我是按照这里改了一下

Collecting package metadata (current_repodata.json): ...working... failed

CondaSSLError: OpenSSL appears to be unavailable on this machine. OpenSSL is required to
download and install packages.

Exception: HTTPSConnectionPool(host='repo.anaconda.com', port=443): Max retries exceeded with url: /pkgs/main/win-64/current_repodata.json (Caused by SSLError("Can't connect to HTTPS URL because the SSL module is not available."))

项目很棒，非常符合我的想法，就是如何支持视频url输入呢？
不是做这方面的，不太懂，大概告诉我思路就行

为啥我的是用cpu做的推理😭，这个怎么修改成gpu呀

环境是win10 ltsc python3.8.
以下是报错信息:
Traceback (most recent call last): sg.run() File "backend/main.py", line 214, in run recognizer = AudioRecogniser(language=self.language) File "backend/main.py", line 32, in __init__ self.model = whisper.load_model(self.model_path) File "D:\Projects\video-subtitle-generator\backend\whisper\__init__.py", line 147, in load_model dims = ModelDimensions(**checkpoint["dims"]) KeyError: 'dims'
medium模型可以正常生成字幕文件.

I built the environment according to doc, and after run python backend/main.py got this error

Traceback (most recent call last):
  File "backend/main.py", line 275, in <module>
    sg.run()                                                                                                                                                                                                                                                  
  File "backend/main.py", line 214, in run                                                                                                                                                                                                                      
    recognizer = AudioRecogniser(language=self.language)                                                                                                                                                                                                      
  File "backend/main.py", line 32, in __init__                                                                                                                                                                                                                  
    self.model = whisper.load_model(self.model_path)                                                                                                                                                                                                          
  File "/home/ddl/video_cut/video-subtitle-generator/backend/whisper/__init__.py", line 149, in load_model                                                                                                                                                      
    model.load_state_dict(checkpoint["model_state_dict"])                                                                                                                                                                                                     
  File "/home/ddl/anaconda3/envs/vsgEnv/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1497, in load_state_dict                                                                                                                                  
    raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(                                                                                                                                                                               
RuntimeError: Error(s) in loading state_dict for Whisper:                                                                                                                                                                                                           
While copying the parameter named "encoder.blocks.0.attn.query.weight", whose dimensions in the model are torch.Size([1024, 1024]) and whose dimensions in the checkpoint are torch.Size([1024, 1024]), an exception occurred : ('CUDA error: no kernel image is available for execution on the device',). 
While copying the parameter named "encoder.blocks.0.attn.key.weight", whose dimensions in the model are torch.Size([1024, 1024]) and whose dimensions in the checkpoint are torch.Size([1024, 1024]), an exception occurred : ('CUDA error: no kernel image is available for execution on the device',).
While copying the parameter named "encoder.blocks.0.attn.value.weight", whose dimensions in the model are torch.Size([1024, 1024]) and whose dimensions in the checkpoint are torch.Size([1024, 1024]), an exception occurred : ('CUDA error: no kernel image is available for execution on the device',). 
While copying the parameter named "encoder.blocks.0.attn.out.weight", whose dimensions in the model are torch.Size([1024, 1024]) and whose dimensions in the checkpoint are torch.Size([1024, 1024]), an exception occurred : ('CUDA error: no kernel image is available for execution on the device',). 

My GPU is 3090, what should I do about it? Looking forward to your reply!

(vsgEnv) E:\video-subtitle-generator\backend>python main.py
Traceback (most recent call last):
File "main.py", line 15, in
import librosa
File "D:\Python\Miniconda3\envs\vsgEnv\lib\site-packages\librosa_init_.py", line 209, in
from . import core
File "D:\Python\Miniconda3\envs\vsgEnv\lib\site-packages\librosa\core_init_.py", line 6, in
from .audio import * # pylint: disable=wildcard-import
File "D:\Python\Miniconda3\envs\vsgEnv\lib\site-packages\librosa\core\audio.py", line 11, in
import scipy.signal
File "D:\Python\Miniconda3\envs\vsgEnv\lib\site-packages\scipy\signal_init_.py", line 331, in
from .peak_finding import *
File "D:\Python\Miniconda3\envs\vsgEnv\lib\site-packages\scipy\signal_peak_finding.py", line 8, in
from scipy.stats import scoreatpercentile
File "D:\Python\Miniconda3\envs\vsgEnv\lib\site-packages\scipy\stats_init.py", line 468, in
from ._rvs_sampling import rvs_ratio_uniforms, NumericalInverseHermite # noqa
File "D:\Python\Miniconda3\envs\vsgEnv\lib\site-packages\scipy\stats_rvs_sampling.py", line 3, in
from ._unuran import unuran_wrapper
File "unuran_wrapper.pyx", line 221, in init scipy.stats._unuran.unuran_wrapper
File "unuran_wrapper.pyx", line 200, in scipy.stats._unuran.unuran_wrapper._setup_unuran
File "messagestream.pyx", line 36, in scipy._lib.messagestream.MessageStream.cinit
OSError: Failed to open file b'C:\Users\\xe5\x8f\xbc\xe7\x9d\x80\xe9\x9b\xb6~1\AppData\Local\Temp\scipy-_60rnr8b'

请输入视频地址: F:\video-subtitle-generator-main\test\test_cn.mp4
return program
'ffmpeg' 不是内部或外部命令，也不是可运行的程序
或批处理文件。
multiprocessing.pool.RemoteTraceback:
"""
Traceback (most recent call last):
File "C:\Users\user007\miniconda3\envs\vsgEnv\lib\multiprocessing\pool.py", line 125, in worker
result = (True, func(*args, **kwds))
File "F:\video-subtitle-generator-main\backend\main.py", line 64, in call
subprocess.check_output(command, stdin=open(os.devnull), shell=use_shell)
File "C:\Users\user007\miniconda3\envs\vsgEnv\lib\subprocess.py", line 415, in check_output
return run(*popenargs, stdout=PIPE, timeout=timeout, check=True,
File "C:\Users\user007\miniconda3\envs\vsgEnv\lib\subprocess.py", line 516, in run
raise CalledProcessError(retcode, process.args,
subprocess.CalledProcessError: Command '['ffmpeg', '-ss', '0.006000000000000005', '-t', '1.524', '-y', '-i', 'C:\Users\user007\AppData\Local\Temp\tmpyt4zyzet.wav', '-loglevel', 'error', 'C:\Users\user007\AppData\Local\Temp\tmpk3bh6dlw.flac']' returned non-zero exit status 1.
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "backend/main.py", line 228, in
sg.run()
File "backend/main.py", line 194, in run
for i, extracted_region in enumerate(pool.imap(converter, regions)):
File "C:\Users\user007\miniconda3\envs\vsgEnv\lib\multiprocessing\pool.py", line 868, in next
raise value
subprocess.CalledProcessError: Command '['ffmpeg', '-ss', '0.006000000000000005', '-t', '1.524', '-y', '-i', 'C:\Users\user007\AppData\Local\Temp\tmpyt4zyzet.wav', '-loglevel', 'error', 'C:\Users\user007\AppData\Local\Temp\tmpk3bh6dlw.flac']' returned non-zero exit status 1.
(vsgEnv) PS F:\video-subtitle-generator-main>

请问是缺了什么组件吗？

Hi friend,

Would you like to tell how to recognize the voice with zh and en output together? thanks

成功打开文件：D:/0software/video-subtitle-generator/test/test_cn.mp4
选择识别的语言: 自动检测
选择识别模式: 标准
['D:/0software/video-subtitle-generator/test/test_cn.mp4']
【处理中】开始生成字幕，此步骤可能花费较长时间，请耐心等待...
Exception in thread Thread-1 (task):
Traceback (most recent call last):
File "C:\Program Files\Python310\lib\threading.py", line 1016, in _bootstrap_inner
self.run()
File "C:\Program Files\Python310\lib\threading.py", line 953, in run
self._target(*self._args, **self._kwargs)
File "D:\0software\video-subtitle-generator\gui.py", line 175, in task
self.sg.run()
File "D:\0software\video-subtitle-generator\backend\main.py", line 227, in run
transcript = recognizer(data)
File "D:\0software\video-subtitle-generator\backend\main.py", line 37, in call
mel = whisper.log_mel_spectrogram(audio_data).to(self.model.device)
File "D:\0software\video-subtitle-generator\backend\whisper\audio.py", line 131, in log_mel_spectrogram
audio = torch.from_numpy(audio)
RuntimeError: Numpy is not available

使用了venv，不知道为什么会和系统上装的python有关系

按照说明，运行程序，执行python backend/main.py，会提示No module named 'fsplit'

然后自己单独pip install fsplit安装fsplit，仍然会提示

Traceback (most recent call last):
File "", line 2, in
File "", line 34, in
File "C:\Users\Fisher\AppData\Local\Temp\pip-install-5zl3_dr0\fsplit_45d13962e7b147008eee4fa18d381872\setup.py", line 11, in
from fsplit import version
File "C:\Users\Fisher\AppData\Local\Temp\pip-install-5zl3_dr0\fsplit_45d13962e7b147008eee4fa18d381872\fsplit_init_.py", line 11, in
from info import version # define version variable
ModuleNotFoundError: No module named 'info'
[end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
error: metadata-generation-failed

× Encountered error while generating package metadata.
╰─> See above for output.

note: This is an issue with the package mentioned above, not pip.
hint: See above for details.

请问如何解决

创建虚拟环境后，均按照readme进行，在环境下运行GUI.PY之后出现报错。
Logs如下：

成功打开文件：F:/video2txt/02/002.mp4
['F:/video2txt/02/002.mp4']
Exception in thread Thread-1:
Traceback (most recent call last):
File "F:\Miniconda\envs\vsgEnv_1\lib\threading.py", line 932, in _bootstrap_inner
self.run()
File "F:\Miniconda\envs\vsgEnv_1\lib\threading.py", line 870, in run
self._target(*self._args, **self._kwargs)
File ".\gui.py", line 173, in task
from backend import main
File "F:\video2txt\VSG\video-subtitle-generator\backend\main.py", line 23, in
from backend import config
File "F:\video2txt\VSG\video-subtitle-generator\backend\config.py", line 10, in
from fsplit.filesplit import Filesplit
ModuleNotFoundError: No module named 'fsplit'

尝试了自动/手动单独装fsplit，但是装不上，因为fsplit包里缺少info库

(vsgEnv_1) (vsgEnv_1) PS F:\video2txt\VSG\video-subtitle-generator> python "E:\Download of Chrome\dist\fsplit-1.0.0\setup.py"
Traceback (most recent call last):
File "E:\Download of Chrome\dist\fsplit-1.0.0\setup.py", line 11, in
from fsplit import version
File "E:\Download of Chrome\dist\fsplit-1.0.0\fsplit_init_.py", line 11, in
from info import version # define version variable
ModuleNotFoundError: No module named 'info'

是我哪里做错了吗？

运行 python gui.py 的时候出现下面的错误

ModuleNotFoundError: No module named 'utils.formatter'

已经尝试过 pip install utils 安装 utils 可是还是出现错误。

Python 版本 3.8.12

conda 和 pip 安装的步骤都没有报错。
以下是在win11系统下，运行main的信息报错信息。

return program
RuntimeError: module compiled against API version 0xf but this version of numpy is 0xe
ImportError: numpy.core.multiarray failed to import

The above exception was the direct cause of the following exception:

SystemError: <built-in method __contains__ of dict object at 0x00000162E8958BC0> returned a result with an error set

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File ".\main.py", line 228, in <module>
    sg.run()
  File ".\main.py", line 199, in run
    transcript = recognizer(data)
  File ".\main.py", line 35, in __call__
    _, probs = self.model.detect_language(mel)
  File "D:\Miniconda3\lib\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "C:\Users\Administrator\Downloads\video-subtitle-generator-main\video-subtitle-generator-main\backend\whisper\decoding.py", line 56, in detect_language
    language_probs = [
  File "C:\Users\Administrator\Downloads\video-subtitle-generator-main\video-subtitle-generator-main\backend\whisper\decoding.py", line 59, in <listcomp>
    for j, c in zip(tokenizer.all_language_tokens, tokenizer.all_language_codes)
  File "C:\Users\Administrator\Downloads\video-subtitle-generator-main\video-subtitle-generator-main\backend\whisper\tokenizer.py", line 228, in all_language_codes
    return tuple(self.decode([l]).strip("<|>") for l in self.all_language_tokens)
  File "C:\Users\Administrator\Downloads\video-subtitle-generator-main\video-subtitle-generator-main\backend\whisper\tokenizer.py", line 228, in <genexpr>
    return tuple(self.decode([l]).strip("<|>") for l in self.all_language_tokens)
  File "C:\Users\Administrator\Downloads\video-subtitle-generator-main\video-subtitle-generator-main\backend\whisper\tokenizer.py", line 141, in decode
    return self.tokenizer.decode(token_ids, **kwargs)
  File "D:\Miniconda3\lib\site-packages\transformers\tokenization_utils_base.py", line 3474, in decode
    token_ids = to_py_obj(token_ids)
  File "D:\Miniconda3\lib\site-packages\transformers\utils\generic.py", line 174, in to_py_obj
    return [to_py_obj(o) for o in obj]
  File "D:\Miniconda3\lib\site-packages\transformers\utils\generic.py", line 174, in <listcomp>
    return [to_py_obj(o) for o in obj]
  File "D:\Miniconda3\lib\site-packages\transformers\utils\generic.py", line 175, in to_py_obj
    elif is_tf_tensor(obj):
  File "D:\Miniconda3\lib\site-packages\transformers\utils\generic.py", line 151, in is_tf_tensor
    return False if not is_tf_available() else _is_tensorflow(x)
  File "D:\Miniconda3\lib\site-packages\transformers\utils\generic.py", line 142, in _is_tensorflow
    import tensorflow as tf
  File "D:\Miniconda3\lib\site-packages\tensorflow\__init__.py", line 37, in <module>
    from tensorflow.python.tools import module_util as _module_util
  File "D:\Miniconda3\lib\site-packages\tensorflow\python\__init__.py", line 37, in <module>
    from tensorflow.python.eager import context
  File "D:\Miniconda3\lib\site-packages\tensorflow\python\eager\context.py", line 33, in <module>
    from tensorflow.python.client import pywrap_tf_session
  File "D:\Miniconda3\lib\site-packages\tensorflow\python\client\pywrap_tf_session.py", line 19, in <module>
    from tensorflow.python.client._pywrap_tf_session import *
ImportError: initialization failed

请问有模型训练的代码吗？

代码调用：
# 1.指定音视频文件路径
wav_path = './test/test.flv'
# 2. 新建字幕提取器
sg = SubtitleGenerator(wav_path)
# 3. 运行字幕生成
ret = sg.run()

While copying the parameter named "decoder.blocks.23.mlp.2.weight", whose dimensions in the model are torch.Size([1024, 4096]) and whose dimensions in the checkpoint are torch.Size([1024, 4096]), an exception occurred : ('CUDA error: no kernel image is available for execution on the device\nCUDA kernel errors might be asynchronously reported at some other API call,so the stacktrace below might be incorrect.\nFor debugging consider passing CUDA_LAUNCH_BLOCKING=1.',).

在Colab上使用Mode = large，生成速度很慢是怎么回事？

GPU 好像 没什么利用到. 建字幕 不用 GPU 的是吗?

貌似解决了: 用管理员权限打开CMD运行就好了。

原问题:

第一次打开, 选择了快速, 能正常开始生成字幕。但当时心急, 看到有错误, 想着直接关掉重新用中等去运行。
结果关掉的时候看到命令行处好像出现了点异常, 之后强行关掉命令行窗口。

之后无论用各种方法, 虽然能打开gui界面, 也能正常打开视频, 但是点运行就只会卡死在那里, 显卡显存也完全没有占用。

而且有个莫名其妙的副作用, 现在我的trojan-Qt5(梯子)启动的时候各种提示端口错误, 之前从未遇到过, 莫名其妙。

目前尝试过把项目重新解压缩, 重启电脑, 都不行。。

Collecting appdirs==1.4.4
Using cached appdirs-1.4.4-py2.py3-none-any.whl (9.6 kB)
Collecting audioread==2.1.9
Using cached audioread-2.1.9.tar.gz (377 kB)
Preparing metadata (setup.py) ... done
Collecting certifi==2021.10.8
Using cached certifi-2021.10.8-py2.py3-none-any.whl (149 kB)
Collecting cffi==1.15.0
Using cached cffi-1.15.0.tar.gz (484 kB)
Preparing metadata (setup.py) ... done
Collecting chardet==4.0.0
Using cached chardet-4.0.0-py2.py3-none-any.whl (178 kB)
Collecting click==8.1.3
Using cached click-8.1.3-py3-none-any.whl (96 kB)
Collecting decorator==5.1.1
Using cached decorator-5.1.1-py3-none-any.whl (9.1 kB)
Collecting filelock==3.6.0
Using cached filelock-3.6.0-py3-none-any.whl (10.0 kB)
Collecting filesplit==3.0.2
Using cached filesplit-3.0.2.tar.gz (5.7 kB)
Preparing metadata (setup.py) ... done
Collecting idna==3.3
Using cached idna-3.3-py3-none-any.whl (61 kB)
Collecting joblib==1.1.0
Using cached joblib-1.1.0-py2.py3-none-any.whl (306 kB)
Collecting librosa==0.9.1
Using cached librosa-0.9.1-py3-none-any.whl (213 kB)
ERROR: Ignored the following versions that require a different python version: 0.36.0 Requires-Python >=3.6,<3.10; 0.37.0 Requires-Python >=3.7,<3.10; 0.38.0 Requires-Python >=3.7,<3.11; 0.38.1 Requires-Python >=3.7,<3.11
ERROR: Could not find a version that satisfies the requirement llvmlite==0.38.0 (from versions: 0.2.0, 0.2.1, 0.2.2, 0.4.0, 0.5.0, 0.6.0, 0.7.0, 0.8.0, 0.9.0, 0.10.0, 0.11.0, 0.12.0.1, 0.12.1, 0.13.0, 0.14.0, 0.15.0, 0.16.0, 0.17.0, 0.17.1, 0.18.0, 0.19.0, 0.20.0, 0.21.0, 0.22.0, 0.23.0, 0.23.2, 0.24.0, 0.25.0, 0.26.0, 0.27.0, 0.27.1, 0.28.0, 0.29.0, 0.30.0, 0.31.0, 0.32.0, 0.32.1, 0.33.0, 0.34.0, 0.35.0, 0.39.0, 0.39.1)
ERROR: No matching distribution found for llvmlite==0.38.0

前面运行的都很顺利，我也不知道这是怎么回事，希望大佬解答