Re:New study: Google manipulates users into constant tracking
by TeMPOraL on Friday February 01, 2019 @04:25PM
(#19058802)
Can confirm. Microsoft Speech API of ~2007 with little training worked very well for me, even across the room on a crappy microphone I hand-soldered to a 2m cable. If that worked fully off-line without a big performance penalty on mid-end 2007 hardware, it surely can work now.
Re:New study: Google manipulates users into constant tracking
by jayd16 on Friday February 01, 2019 @07:53PM
(#19060214)
If its that easy you should make an app because there's certainly a market for it. The reality is online connected services are far far better.
There's a reason voice rec has only now finally stuck.
Re:New study: Google manipulates users into constant tracking
by ben_w on Friday February 01, 2019 @08:14PM
(#19060305)
On my iPhone, the voice keyboard works pretty well in airplane mode. Not quite as well as when it has signal, but the difference is small… with the sole exception that it recognises everything as if it is English in Airplane mode, and only uses the language of the keyboard (if different) when I have signal.
Re:New study: Google manipulates users into constant tracking
by tinus_hn on Friday February 01, 2019 @09:17PM
(#19060588)
By default, iOS dictation is also online.
Voice rec was local until about ten years ago. You could even buy software dictation products that installed from CDROM and didn't need a network connection.
I shouldn't need a network connection to do anything voice on my own device. It should only reach out to the network for things it doesn't have local knowledge of.