search

MiteshPuthran / Speech-Emotion-Analyzer

The neural network model is capable of detecting five different male/female emotions from audio speeches. (Deep Learning, NLP, Python)
MIT License
1.29k stars 436 forks source link

how to run this code in flask server #23

Closed Nikhilbharadwaj08 closed 5 years ago

Nikhilbharadwaj08 commented 5 years ago

Hi How do i run your code in flask server,

MiteshPuthran commented 5 years ago

@Nikhilbharu08, you can used the trained model directly or use your own model.

Just create your server code and front end which you can find easy tutorials online. Include your trained model location and supply it your input. Care should be taken are regarding the audio clippings with the desired sampling rate. If they don't match the model wouldn't make any predictions.

Nikhilbharadwaj08 commented 5 years ago

@Nikhilbharu08, you can used the trained model directly or use your own model.

Just create your server code and front end which you can find easy tutorials online. Include your trained model location and supply it your input. Care should be taken are regarding the audio clippings with the desired sampling rate. If they don't match the model wouldn't make any predictions.

yeah i have created a front end to choose a audio file of .wav format file,
can you please give more details/info regarding the sampling rate which you mentioned !!!?

MiteshPuthran commented 5 years ago

Hey @Nikhilbharu08, sampling rate is what you change according to the number of features you want to extract from the audio files. Higher sampling rates means more features.