Kerolos ghobrial

some cases the model can not produce any outputs.

1

I trained a model based on different English Database. when i run to predict phonemes of my test Database. In some words (Graphemes), the model stuck and can not produce...

is there anyway to use a word lexicon dictionary during the decoding?, i wanna make a link (map) between the output from CTC to the input in the Language model...

streaming_convnets model. the WER does not decrease during training (it remains stable at 100% for dev and train)

6

I tried to use the streaming_convnets from recipes: i used two different lexicon and token(character based model vs wordpiece based model): I used for that a small dataset for training...

How to decode by using a pre-trained acoustic model and customized 3-gram LM

10

### i would like to use the pre-trained model which has been trained in the Librispeech dataset into my own customized data, with changing only the language model and the...

using CTC loss instead of Cross Entropy loss:

1

In all Tutorials, the cross entropy has been used. i would like to change this loss function to be CTC (i am using those model for g2p purpose not for...

adding a function to pass (image as cv::Mat format from cpp) and receive numpy variable in python script

1

Dear Mr Wei Li(Lee); i would like to ask you if it is possible to add a function to convert cv::mat to numpy and path this numpy variable to a...

How can labeling my own dataset

4

How can I label a customized dataset to be able to use in Caffe_segnet? Which tool or program I should use to create my own label dataset with the same...

Hello Sir, Does this Library provide a synchronization between multi-IMU sensors? "in real time" if yes How can i do that? thanks in advance.

GMM-HMM (Cross-validation)

1

Dear Mr.Qiuqiang Kong, I would like to do Cross Validation to choose the best parameters such as # of states and # of observation. -My data is 9 dimensions because...