On 07/18/2015 03:42 AM, Eric wrote:
On 07/17/2015 05:25 PM, charles meyer wrote:
For those on the list using or contemplating Linux, this was
interesting.
http://www.techrepublic.com/article/alternative-is-the-question-linux-is-the-answer/
No one really addressed software only written for Windows - e.g.
Dragon Naturally Speaking.
There doesn't seem to be a good open source or Linux-based voice
recognition software.
That can be a deal breaker.
yes it can be. As someone dependent on NaturallySpeaking, I can tell
you that most of the native limits speech recognition solutions do not
measure up and would probably take a multimillion dollar effort to
make a functional equivalent.
Then there is the issue of supporting it, fixing bugs, adding
features, working with the community current open-source funding
techniques just don't measure up to the financial needs of critical
projects like this would be.
So how is it, I am using NaturallySpeaking to dictate on the next?
Well the wine solution is pretty fragile and doesn't work with the
latest NaturallySpeaking. It still has problems with using community
supported extensions such as natlink and vocola.
Instead, I'm using a different solution which is running
NaturallySpeaking in a Windows virtual machine and using a tool I
developed called speech bridge (See github) to link speech
recognition functionality in Windows to whatever the hell Linux is doing.
It's moving slow, I am changing things as I need them. I'm getting to
the point where I could use some assistance but I'm going to hold off
asking until I have a good specification/Definition.
--- eric
There are only two or three Win-only software I need to use, that I have
not found an easy to use alternative for a .deb based Linux OS. Also, I
have not found any way to use my external USB-based wifi antenna, for
"better/faster" useful access 10 floors below my router's antennas. IF
so, then Linux would be use for that work "environment",
Also, for the "speech issue", I have never been able to got that to work
properly for Ubuntu and Linux Mint, as least text to speech. I have not
worked with speech to text or speech "system control" before, so I have
not really looked into that, much.