First of all, based on my understanding of Mycroft architecture, voice recognition is done on the device itself, as well as application selection.
Hence, the voice command AND the command logic is done on the mobile device, and DO NOT access a remote computer.
Also, the prior art is likely very strong as evidenced in the patent itself:
"from a mobile device to remotely access and control a
computer are known in the art. However, such prior art
systems are application-specific, meaning they are configured to allow the person to use voice commands from a
mobile device to remotely access and control a specific
application at a computer. Therefore, the prior art systems 25
require the person to have multiple mobile devices and/or
systems to remotely access and control the different applications at a computer. Additionally, the prior art systems
limit the audible and visible feedback the person can receive
from a computer while using voice commands from a 30
mobile device to remotely access and control the computer. "
So the patent admits that prior art exists for sending commands to specific applications, but not for general application? , I fail to see the difference.
All voice commands are sent for a specific application.
Hence, since any voice command is for a specific applications, I fail to see how Mycroft violated the current claim.
If you want an example of the prior art, here is a very famous system (from 2006)
In general, the patent shows its age, such that it confines itself to a simple client-server architecture, where the mobile device gets the audio, and the remote computer (the server) does speech recognition and command selection.
However, if you do the speech recognition on the mobile device, as well as the app selection, I think that the patent is no longer valid.
First of all, based on my understanding of Mycroft architecture, voice recognition is done on the device itself, as well as application selection.
Hence, the voice command AND the command logic is done on the mobile device, and DO NOT access a remote computer.
Also, the prior art is likely very strong as evidenced in the patent itself:
"from a mobile device to remotely access and control a computer are known in the art. However, such prior art systems are application-specific, meaning they are configured to allow the person to use voice commands from a mobile device to remotely access and control a specific application at a computer. Therefore, the prior art systems 25 require the person to have multiple mobile devices and/or systems to remotely access and control the different applications at a computer. Additionally, the prior art systems limit the audible and visible feedback the person can receive from a computer while using voice commands from a 30 mobile device to remotely access and control the computer. "
So the patent admits that prior art exists for sending commands to specific applications, but not for general application? , I fail to see the difference.
All voice commands are sent for a specific application.
Hence, since any voice command is for a specific applications, I fail to see how Mycroft violated the current claim.
If you want an example of the prior art, here is a very famous system (from 2006)
http://www.speech.cs.cmu.edu/letsgo/
In general, the patent shows its age, such that it confines itself to a simple client-server architecture, where the mobile device gets the audio, and the remote computer (the server) does speech recognition and command selection.
However, if you do the speech recognition on the mobile device, as well as the app selection, I think that the patent is no longer valid.