24:30 I would try on various epochs just to be sure, because if training it on more epochs would result in loss of the pretrained data then no training should be done. Good tutorial none the less 👍
I am a student trying to learn ML. Can you please show my how to use this trained model for prediction using yhat = model.predict(). I see the input X is of shape 5600,40 but when I try to predict for a new 3 second audio file, the np array default shape is 66150 and I am getting Invalid input shape error
you can run the model using jupyter notebook by installing anaconda software in your machine. please check my initial videos in the playlist to get a clear idea.
A blessed day sir. I would like to ask how do we saved and download the model for future predictions? So that we won't rerun the entire code anymore. I would also like to ask how do we load the saved and downloaded model on other notebooks to test if it can generate prediction?
@@HackersRealm I am an engineer, Can you also tell me if we get speaker attributions like gender or age with this wave2 transformer.? Please provide any article or link for reference
@@jitindoriya2375 not really sure whether you can extract these attributes, if you find a pretrained weights of these specific applications, you could use that to get that information
24:30 I would try on various epochs just to be sure, because if training it on more epochs would result in loss of the pretrained data then no training should be done. Good tutorial none the less 👍
you tested patience level bro 😭😭😭😭😭
are there any models that give the classification along with the arousal and valence values?
I am a student trying to learn ML. Can you please show my how to use this trained model for prediction using yhat = model.predict(). I see the input X is of shape 5600,40 but when I try to predict for a new 3 second audio file, the np array default shape is 66150 and I am getting Invalid input shape error
I am a beginner can u help me , exactly how I can use your model in my machine
you can run the model using jupyter notebook by installing anaconda software in your machine. please check my initial videos in the playlist to get a clear idea.
A blessed day sir.
I would like to ask how do we saved and download the model for future predictions?
So that we won't rerun the entire code anymore.
I would also like to ask how do we load the saved and downloaded model on other notebooks to test if it can generate prediction?
www.tensorflow.org/tutorials/keras/save_and_load
please refer this link for complete detail
Can you define how you upload audio_path and labels
How can I use it predict for single audio file ?
@@jitindoriya2375 i have done it at the end of the video
@@HackersRealm I am an engineer, Can you also tell me if we get speaker attributions like gender or age with this wave2 transformer.? Please provide any article or link for reference
We have developed Vocie AI assistant but it was missing emotion so trying with this method.
@@jitindoriya2375 not really sure whether you can extract these attributes, if you find a pretrained weights of these specific applications, you could use that to get that information
Could you please add how can I test the model with a new audio file?
I think I have added at the end, you have to give the path of the audio file and get the prediction.
the code link which you are providing is not accable ,please uplaod the code again
I have checked now, it's working without issues. May I know what issue you're facing?
Make .h5 file deploy it in app. How to do that
I'm trying to develop a speech aggression level detection model. How can I use this to do that?? Pleasee help me!!
do you have the dataset similar to the same structure like in the video?
@@HackersRealm yes..
How can I use only one single data only one audio sir
@@lifterboy9875 what do you mean by single data