I've been using DragonNaturallySpeaking and it's various hooks for a little while now (my setup is a much smaller set of : https://youtu.be/8SkdfdXWYaI), but lately the tools have degraded and support is hit or miss at best. I was hoping to use CMUSphinx and some other java libraries to have a bit more "stability" in my "assistant."
I have fiddled in a very limited capacity with the provided examples and read through the documentation and articles. Before I invest too heavily I was hoping to get some clarification on:
"For command and control, it was popular to use a finite state grammar for a long time. Unfortunately, we could not recommend that to you now days. It’s way better to employ a medium vocabulary recognizer with semantic analysis framework on the top to improve user experience and let him use more or less natural language. In short, don’t build command and control, build intelligent assistants instead."
The number of words I need to recognize (for the way I use it) number ~100. I don't use voice control for dictation I use it for very specific IDE-centric commands (run/copy/paste/complete/follow/back) and then use the keyboard to complete where neccessary. I do however need it to recognize rather quickly to fit in my workflow - it's worth noting I work in a very low noise environment.
Is Sphinx an appropriate library for this?
TLDR: I need rapid recongition with a very small word set - will Sphinx fu-fill these needs?
Thanks!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi All!
Backstory:
I've been using DragonNaturallySpeaking and it's various hooks for a little while now (my setup is a much smaller set of : https://youtu.be/8SkdfdXWYaI), but lately the tools have degraded and support is hit or miss at best. I was hoping to use CMUSphinx and some other java libraries to have a bit more "stability" in my "assistant."
I have fiddled in a very limited capacity with the provided examples and read through the documentation and articles. Before I invest too heavily I was hoping to get some clarification on:
"For command and control, it was popular to use a finite state grammar for a long time. Unfortunately, we could not recommend that to you now days. It’s way better to employ a medium vocabulary recognizer with semantic analysis framework on the top to improve user experience and let him use more or less natural language. In short, don’t build command and control, build intelligent assistants instead."
From the doc here: https://cmusphinx.github.io/wiki/tutorialbeforestart/
The number of words I need to recognize (for the way I use it) number ~100. I don't use voice control for dictation I use it for very specific IDE-centric commands (run/copy/paste/complete/follow/back) and then use the keyboard to complete where neccessary. I do however need it to recognize rather quickly to fit in my workflow - it's worth noting I work in a very low noise environment.
Is Sphinx an appropriate library for this?
TLDR: I need rapid recongition with a very small word set - will Sphinx fu-fill these needs?
Thanks!
Yes
Awesome! Thank you!
You are welcome