Not Running speechEmotionRecognition.py

[lenoirdevinci]

When I try to run the app, I have this error : Using TensorFlow backend. /usr/local/lib/python3.6/site-packages/pydub/utils.py:170: RuntimeWarning: Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work warn("Couldn't find ffmpeg or avconv - defaulting to ffmpeg, but may not work", RuntimeWarning) Traceback (most recent call last): File "./src/speechEmotionRecognition.py", line 21, in model = tf.keras.models.load_model('model/lstm_cnn_rectangular_lowdropout_trainedoncustomdata.h5') File "/usr/local/lib/python3.7/site-packages/tensorflow_core/python/keras/saving/save.py", line 146, in load_model loader_impl.parse_saved_model(filepath) File "/usr/local/lib/python3.7/site-packages/tensorflow_core/python/saved_model/loader_impl.py", line 83, in parse_saved_model constants.SAVED_MODEL_FILENAME_PB)) OSError: SavedModel file does not exist at: model/lstm_cnn_rectangular_lowdropout_trainedoncustomdata.h5/{saved_model.pbtxt|saved_model.pb}

Can you help me ?

[SuyashMore]

src/speechEmotionRecognition.py , line 21, model = keras.models.load_model('model/lstm_cnn_rectangular_lowdropout_trainedoncustomdata.h5')

maybe try using the absolute path of the file on your system.

[lenoirdevinci]

Thanks for your quick response. And congratulation for your work. I already tried multiple solutions :

• absolute path
• move h5 file in the same location as src/speechEmotionRecognition.py
• use \ as oposite / No solution found

file_name = os.path.dirname(__file__) +'\\model\\lstm_cnn_rectangular_lowdropout_trainedoncustomdata.h5' model = tf.keras.models.load_model(file_name)

model = tf.keras.models.load_model('model/lstm_cnn_rectangular_lowdropout_trainedoncustomdata.h5')

model = tf.keras.models.load_model('lstm_cnn_rectangular_lowdropout_trainedoncustomdata.h5')

[SuyashMore]

Let's try some Debugging.

try Creating a New file just loading the model

import keras
model = keras.models.load_model('model/lstm_cnn_rectangular_lowdropout_trainedoncustomdata.h5')
print(model.summary())

• if it doesn't work , try loading other models present in /src/model directory.
• problems seems to be with the path I reckon (I was successfully able to run the code again on my system a few minutes ago). try going inside /src running the code

  $ cd src/
  $ python speechEmotionRecognition.py

[lenoirdevinci]

Hello, try Creating a New file just loading the model. I don't know how to do this

if it doesn't work , try loading other models present in /src/model directory. Do not work

$ cd src/ $ python speechEmotionRecognition.py. Ok this is working. When I launch from src/ I passed a step. I don't the same issue anymore. Now I have a GPU issue.

toolkits.initialize_GPU(args) AttributeError: module 'toolkits' has no attribute 'initialize_GPU'

I tried to install Cuda and Nvidia drivers but no success. Is it a possibility into your app to change GPU by CPU ? I didn't where I can do this. Or may be you think about another solution

[lenoirdevinci]

Ok when I put in comment the line #toolkits.initialize_GPU(args), it seems to work without GPU. But now I don't see the CSV generated. Any idea ?

[SuyashMore]

 with open('SER_'+subdir+'.csv', 'w') as csvFile:
            writer = csv.writer(csvFile)
            for i in range(len(filenames)):
                csvData = [filenames[i], 'person01',predictions[i][0],'person02',predictions[i][1]]
                print("filename:",filenames[i],",Predicted Emotion := Person1:",predictions[i][0],",Person2:",predictions[i][1])
                writer.writerow(csvData)
        csvFile.close()

check out .csv files starting with "SER_"

[lenoirdevinci]

Yes I saw this part of the code but I don't know why it isn't working. And I have no errors displayed

[chinmaynehate]

Hey @lenoirdevinci,

I think there are some dependency issues. With the last commit, we should have fixed it. Try cloning the master branch and repeat the dependency installation process with the updated setup.sh again. (We have also updated the README.md. Please refer to it if you have any doubts about installing dependencies and running the code.) Let us know if you have any other problems.

Best, Chinmay

[lenoirdevinci]

Thank you a lot. I will try this quickly

[SuyashMore]

Closing Issue for now. Let us know if you face any other issues!

Regards, Suyash