Hi,
I am diagnosed with ‘Wrist tendinitis’/’tenosynovitis’ due to RSI (Repetitive Strain Injury). Hence I started using speech recognition as much as possible. On windows, controls are 80% accurate and dictation is 50% accurate. With practice, I am trying to use keyboard and voice recognition just the same way a pianist would sing.
I think with the wealth of research in speech recognition, it is possible to get a decent accuracy
1. for coding (UVM, systemverilog) as there are predefined set of keywords, classes, functions etc.
2. for GUI tool controls
An example of coding:
to see in editor "class myenv extends uvm_env;"
I would
Say: "class" type: myenv Say:"extends uvm_env;"
For GUI usage, tool vendors/users could come up with a number of useful features similar to keyboard shortcuts. These features may be implemented both on Windows and Linux platforms.
(Hope in few years, not only myself, all engineers are able to sing their code into a tool and debug it too!)
Thanks for reading my request!
Regards,
Kal Gandikota
P.S.: I used windows speech recognition to write this e-mail as I am suffering from wrist pain. Please kindly ignore grammar mistakes.