dessa-oss / fake-voice-detection Goto Github PK

I tried to get it working, but there were way too many problems. It isn't in development or updated anymore so I was just wondering if this should still be working.

If it should still be working through the steps given then I'll give it a try again (reinstall everything through the steps). But I feel like it's just a broken repository at the moment.

Hey,
I was trying to use the model and some issues came up.

• The first challenge I faced is that the requirements.txt file is missing, therefore the manual package installation was a little bit tricky sometimes (newest version of tensorflow does not work for instance)
• The second and bigger problem is that I tried the forward test by using the pre-trained model. I followed the instructions (copied my audio into the right folder, run the inference.py), however the prediction for both fake and real recordings were really low values. I even tried out the pre-trained model with your real voice data and it still gave only a probability of 0.3% of the real clip is being real which is supposed to be close to 100%...

1. The only error I got is: bash: scp_spectrogram.sh: No such file or directory
2. Also I'm not running the Atlas for trying out the pre-trained model which I'm not sure is correct or not.

Do you have any suggestion what do I miss/should install, modify to get more reasonable probabilities?
Thanks!

I wish to train on my own dataset which consists of real and fake wav files. May I know how I can do so in terms of preprocessing and tuning of the hyperparameters?

Hi All,
I did not have NVIDIA GPU on my local machine. I am trying to run the code on Intel GPU but did not got success.

I wanted to confirm that can I run this code on Intel GPU or not ?

Many Thanks,
Kinjal

At test/inference time, what's the maximum length of the audio array that can be tested? I'm having trouble finding this out from the code.

If an audio array is longer than this max_length is trimmed?

We are trying to run https://github.com/dessa-public/fake-voice-detection and got the following error and the process is getting killed. Not sure if you could help with this.(fake_voice)

neutigers@tensorbook:~/Desktop/fake_voice_detection/fake-voice-detection/code$ python main.py
WARNING:tensorflow:
The TensorFlow contrib module will not be included in TensorFlow 2.0.
For more information, please see:
  * https://github.com/tensorflow/community/blob/master/rfcs/20180907-contrib-sunset.md
  * https://github.com/tensorflow/addons
  * https://github.com/tensorflow/io (for I/O related ops)
If you depend on functionality not listed there, please file an issue.Using TensorFlow backend.
/home/neutigers/anaconda3/envs/fake_voice/lib/python3.7/site-packages/sklearn/externals/joblib/__init__.py:15: DeprecationWarning: sklearn.externals.joblib is deprecated in 0.21 and will be removed in 0.23. Please import this functionality directly from joblib, which can be installed with: pip install joblib. If this warning is raised when loading pickled models, you may need to re-serialize those models with scikit-learn 0.21+.
  warnings.warn(msg, category=DeprecationWarning)
/home/neutigers/anaconda3/envs/fake_voice/lib/python3.7/site-packages/foundations_contrib/config_manager.py:90: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
  return yaml.load(file)
/home/neutigers/anaconda3/envs/fake_voice/lib/python3.7/site-packages/foundations_contrib/cli/config_listing.py:41: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
  return yaml.load(file.read())
/home/neutigers/anaconda3/envs/fake_voice/lib/python3.7/site-packages/foundations_internal/config/execution.py:18: YAMLLoadWarning: calling yaml.load() without Loader=... is deprecated, as the default Loader is unsafe. Please read https://msg.pyyaml.org/load for full details.
  schema = yaml.load(file.read())
loading the preprocessed training, validation and test data.
Killed
GitHubGitHub
dessa-public/fake-voice-detection
Using temporal convolution to detect Audio Deepfakes - dessa-public/fake-voice-detection

Hi, I appreciate your work

Is there any way to run the model on google colab? (Or any online tool) since my machine does not fulfill the specs.

(env) C:\Users\ram\Desktop\Foundation_Atlas\env>atlas-server start
2020-03-14 11:26:01,607 - Foundations Atlas Server - INFO - Starting atlas-ce-tracker service...
2020-03-14 11:26:02,929 - Foundations Atlas Server - INFO - Started atlas-ce-tracker service
2020-03-14 11:26:03,042 - Foundations Atlas Server - INFO - Starting atlas-ce-archive-server service...
2020-03-14 11:26:04,315 - Foundations Atlas Server - INFO - Started atlas-ce-archive-server service
2020-03-14 11:26:04,447 - Foundations Atlas Server - INFO - Starting atlas-ce-rest-api service...
2020-03-14 11:26:05,584 - Foundations Atlas Server - INFO - Started atlas-ce-rest-api service
2020-03-14 11:26:05,712 - Foundations Atlas Server - INFO - Starting atlas-ce-gui service...
2020-03-14 11:26:06,688 - Foundations Atlas Server - INFO - Started atlas-ce-gui service
2020-03-14 11:26:06,714 - Foundations Atlas Server - ERROR - Cannot start atlas-ce-local-scheduler
2020-03-14 11:26:06,715 - Foundations Atlas Server - ERROR - Please try re-installing Atlas Server via the installer
2020-03-14 11:26:06,720 - Foundations Atlas Server - INFO - Stopping atlas-ce-gui service...
2020-03-14 11:26:17,589 - Foundations Atlas Server - INFO - Stopped atlas-ce-gui service
2020-03-14 11:26:17,591 - Foundations Atlas Server - INFO - Removing atlas-ce-gui service...
2020-03-14 11:26:17,674 - Foundations Atlas Server - INFO - Removed atlas-ce-gui service
2020-03-14 11:26:17,676 - Foundations Atlas Server - INFO - Stopping atlas-ce-rest-api service...
2020-03-14 11:26:19,518 - Foundations Atlas Server - INFO - Stopped atlas-ce-rest-api service
2020-03-14 11:26:19,520 - Foundations Atlas Server - INFO - Removing atlas-ce-rest-api service...
2020-03-14 11:26:19,594 - Foundations Atlas Server - INFO - Removed atlas-ce-rest-api service
2020-03-14 11:26:19,595 - Foundations Atlas Server - INFO - Stopping atlas-ce-archive-server service...
2020-03-14 11:26:20,246 - Foundations Atlas Server - INFO - Stopped atlas-ce-archive-server service
2020-03-14 11:26:20,247 - Foundations Atlas Server - INFO - Removing atlas-ce-archive-server service...
2020-03-14 11:26:20,291 - Foundations Atlas Server - INFO - Removed atlas-ce-archive-server service
2020-03-14 11:26:20,292 - Foundations Atlas Server - INFO - Stopping atlas-ce-tracker service...
2020-03-14 11:26:21,140 - Foundations Atlas Server - INFO - Stopped atlas-ce-tracker service
2020-03-14 11:26:21,141 - Foundations Atlas Server - INFO - Removing atlas-ce-tracker service...
2020-03-14 11:26:21,176 - Foundations Atlas Server - INFO - Removed atlas-ce-tracker service
2020-03-14 11:26:21,176 - Foundations Atlas Server - INFO - Removing foundations-atlas network service...
2020-03-14 11:26:21,308 - Foundations Atlas Server - INFO - Removed foundations-atlas network service...

How to resolve this error in local. Please help me to rid of these error.

Thanks,
Kinjal

Importing the model using tensorflow and keras gives me the following error:

NameError Traceback (most recent call last)
in ()
----> 1 model = tf.keras.models.load_model('model.h5')

12 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/keras/layers/core.py in (x)
680
681 def compute_output_shape(self, input_shape):
--> 682 input_shape = tensor_shape.as_shape(input_shape).as_list()
683 if not input_shape:
684 output_shape = tensor_shape.TensorShape([1])

NameError: name 'customPooling' is not defined

Hello, Can anyone help me understand the inference logic used in the code?
I used the pre-trained model and I can't seem to get correct results. No video is crossing the probability of 10% while classifying any audio file. What does the model prediction actually output?
what does [[0.00045425]] mean?

Hello,

I tried to install the requirements.txt file but shows an error in the line " import numpy as np" saying no module found as "numpy" .

UPDATE:
I tried to manually install all the packages.
I was not able to install lws and PyYAML. The process got terminated.

Kindly help me resolve this issue.

Regards,
Harshit

Hi when I ran the inference.py file with all the audio files in the 'real' subfolder - it misclassified them as 'fake'. I just wanted to check that the pre-trained model is the correct one?

I am not able to download the dataset by running "bash download_data.sh". I have also tried with git bash to download it, but did not got any luck. Attaching trace here:

$ bash download_data.sh download_data.sh: line 1: apt-get: command not found download_data.sh: line 2: apt-get: command not found download_data.sh: line 3: apt-get: command not found download_data.sh: line 4: apt-get: command not found download_data.sh: line 10: wget: command not found unzip: cannot find or open realtalk.zip, realtalk.zip.zip or realtalk.zip.ZIP. mv: cannot stat 'realtalk/*': No such file or directory rm: cannot remove 'realtalk.zip': No such file or directory rm: cannot remove 'realtalk/': No such file or directory download_data.sh: line 20: wget: command not found unzip: cannot find or open preprocessed_data.zip, preprocessed_data.zip.zip or preprocessed_data.zip.ZIP. rm: cannot remove 'preprocessed_data.zip': No such file or directory download_data.sh: line 28: wget: command not found
Can any one please me to get rid of these one.

Many Thanks.
Kinjal