Arseniy Gorin

Sorry @Faizi-AdnanFahad, this project was quite abandoned recently. Unless you want to stick to sphinx decoder, I would suggest looking at https://github.com/alphacep/ros-vosk It should be more accurate and overall better...

Sorry folks, this project was quite abandoned recently. Unless you want to stick to sphinx decoder, I would suggest looking at https://github.com/alphacep/ros-vosk It should be more accurate and overall better...

I believe we had this issue a few years ago and a good speed-up was achieved by pre-computing the cache once and saving it to the file. If you pre-compute...

Which result do you have with distance library? Can you confirm you use keras 1.1.1?

Usually it does. Without it you compute word accuracy, otherwise its more related to character accuracy

Folks say RMSProp might be even better for recurrent networks. I did not compare on this particular toolkit. By the way, a pull request is always welcomed ;)

we will unlikely be able to fix it unless re-write tensorflow

Hi guys. Is my understanding correct that `ds.filter` would only allow downsampling (reducing the number of samples from dominating class)? I was trying to implement upsampling in a view but...

Thanks @farizrahman4u. The solution with copy is clear but I would ideally like to avoid copying the data. I am currently using PyTorch Datasets and DataLoader with weighted sampler -...