End-to-End STT

on

 

End-to-end generally refers to things like CTC, transformers, RNN-T. Generally, anything where you can code the entire thing with PyTorch/TensorFlow without C++ code. but again, End to end is not really a meaningful or important category. Generally it's considered trendy to be end to end.

Kaldi STT probably would not be considered end-to-end STT.

I don't have the sample for the end to end STT. 

RNN-T: Recurrent Neural Network Transducer

CTC: Connectionist temporal classification 

Transformers: A transformer is a deep learning model that adopts the mechanism of self-attention,    differentially weighting the significance of each part of the input data. It is used primarily in the field of natural language processing (NLP)[1] and in computer vision (CV).[2]

https://en.wikipedia.org/wiki/Transformer_(machine_learning_model)


Comments

Post a Comment