Rithesh Kumar

icon indicating a website link ritheshkumar.com icon indicating an email [email protected]

icon company @adobe-research @mila-iqia icon location Toronto icon location Research Scientist @ Adobe Research. Ex Descript Inc., Mila lab

Results 16 comments of Rithesh Kumar

Cannot run the program

Hi, Sorry for the missing response in this thread. I wrote this code when it was tensorflow version (0.9?), and didn't manage to get time with my busy schedule to...

output the whole sequence within TF

I already had this code. i forget to push it, Check if it works now and let me know,

output the whole sequence within TF

@moezlinlin @mingchen62 I'm sorry i'm quite busy, i don't get the time to fix bugs everytime tensorflow updates the version. Please try fixing it yourself, i'm happy to accept your...

output the whole sequence within TF

> > @moezlinlin @mingchen62 I'm sorry i'm quite busy, i don't get the time to fix bugs everytime tensorflow updates the version. Please try fixing it yourself, i'm happy to...

Details on training and inferencing the 16kHz variant

This is the config for training the 16 KHz model https://github.com/descriptinc/descript-audio-codec/blob/main/conf/final/16khz.yml The input is always resampled to 16 KHz and output will always match the input sampling rate. The documentation...

Question regarding kernel size and stride direction in the MRD module

Thanks for bringing this to our notice. This was an unintentional error when porting the code from the original implementation of Multi-resolution Spectrogram Discriminator (MRSD) from UnivNet. You're right about...