Sreyan Ghosh

icon indicating a website link https://sreyan88.github.io/ icon indicating an email [email protected]

icon company University of Maryland, Ex-Nvidia & Cisco, MIDAS@IIIT-D, Speech Lab@IIT-M icon location College Park, Maryland icon location Ph.D. in CS @ University of Maryland College Park, Speech, Audio and Language Processing Researcher

Results 44 comments of Sreyan Ghosh

Training and Testing code

Hi @JuanFMontesinos , I have two questions and would be grateful if you could answer them! (1) Can you please provide me with the code for: complex_division() (2) I want...

Training and Testing code

Thank You so much for your reply! Can you also please help me with your stft parameters?

when does the error 'no utterance found' arrises ? can you please explain it.

Same, is this problem solved?

Key to checkpoints in drive

Hi @RetroCirce , Thank You for the reply. Are they all tiny models or base/large in the AudioSet folder and what is the sampling rate you used for them? Additionally,...

Genius-f logs for NER

I try to fine-tune with Genius-f on a custom NER dataset and it seems to converge in about 2 epochs (lowest eval loss). Is this behavior normal?

Some bug fixes

Hi @VlaDanilov , Thank You for the suggestions. Would you be willing to create a pull request please? If not, I can push later! Thank You!

How to train and validate the model?

Hello! Thank You for your interest. You can follow the steps in our repo. Also, we will soon update our repo with some changes by end of the week so...

How to train and validate the model?

@Utkarsh4430 can you please update our latest code with instructions? Thank You!

How to train and validate the model?

Hi @Coding511 , We will release our latest code with better results in 1-2 days. Thank You for your patience!

How to train and validate the model?

Extremely sorry for the delay in response. Our paper has been accepted to InterSpeech 2023, and we have pushed new code with improved performance. Please let us know of any...