Reply to thread

Message

hey there, glad to see more enhancements
[USER=17179]Bram[/USER] -for the issue of the PC responding to regular conversations, this is something that will happen with a room mounted mic vs one on your person.
A few things I have done to improve this
- pick an activation phrase that is pretty unique. It's best to give the PC a name, since "computer" can come up in conversation.
- spend a lot of time training the PC and DEFINATELY make sure "adaptive learning" is turned off in the MS speech settings. This is one time where you don't want the PC to be learning by listening to the TV.
- set the time out time to be quite short. I set mine for four seconds, which is short enough not to be a problem with mistaken activations.
- make sure the activation phrase has a plus in front of it. This signifies a more certain signal is needed to activate the engine.

One feature request I have:
It's pretty gimmicky and may be hard with the MS speech engine, but is it possible to have a command where the PC will repeat what you say? for example, you can say "Say hello to XXX" and the PC will then say "Hello XXX".

Also, the voice I am now using is Callie from a company called cepstral. IT's actually very lifelike and only cost US$29. I may have already said this, sorry if I did

and for a really off the wall request - Do you know of any way to access the voice engine remotely?
If my housemate is watching TV, I'd like to be able to access a service from a different PC that will let me type in something for the PC to say. I figure I could have loads of fun with something like that. I'm guessing the only way would be to use a web service though. Any ideas?

I'm in Mexico at the moment, but will load up the new version when I get home.

<blockquote data-quote="Marcusb" data-source="post: 175293" data-attributes="member: 11550">hey there, glad to see more enhancements <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite1" alt=":)" title="Smile&nbsp; &nbsp; :)" loading="lazy" data-shortname=":)" />[USER=17179]Bram[/USER] -for the issue of the PC responding to regular conversations, this is something that will happen with a room mounted mic vs one on your person.A few things I have done to improve this - pick an activation phrase that is pretty unique. It's best to give the PC a name, since &quot;computer&quot; can come up in conversation.- spend a lot of time training the PC and DEFINATELY make sure &quot;adaptive learning&quot; is turned off in the MS speech settings. This is one time where you don't want the PC to be learning by listening to the TV.- set the time out time to be quite short. I set mine for four seconds, which is short enough not to be a problem with mistaken activations.- make sure the activation phrase has a plus in front of it. This signifies a more certain signal is needed to activate the engine.One feature request I have:It's pretty gimmicky and may be hard with the MS speech engine, but is it possible to have a command where the PC will repeat what you say? for example, you can say &quot;Say hello to XXX&quot; and the PC will then say &quot;Hello XXX&quot;.Also, the voice I am now using is Callie from a company called cepstral. IT's actually very lifelike and only cost US$29. I may have already said this, sorry if I did <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite1" alt=":)" title="Smile&nbsp; &nbsp; :)" loading="lazy" data-shortname=":)" />and for a really off the wall request <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite1" alt=":)" title="Smile&nbsp; &nbsp; :)" loading="lazy" data-shortname=":)" /> - Do you know of any way to access the voice engine remotely?If my housemate is watching TV, I'd like to be able to access a service from a different PC that will let me type in something for the PC to say. I figure I could have loads of fun with something like that.&nbsp; I'm guessing the only way would be to use a web service though.&nbsp; Any ideas?I'm in Mexico at the moment, but will load up the new version when I get home.</blockquote>

[QUOTE="Marcusb, post: 175293, member: 11550"] hey there, glad to see more enhancements :) [USER=17179]Bram[/USER] -for the issue of the PC responding to regular conversations, this is something that will happen with a room mounted mic vs one on your person. A few things I have done to improve this - pick an activation phrase that is pretty unique. It's best to give the PC a name, since "computer" can come up in conversation. - spend a lot of time training the PC and DEFINATELY make sure "adaptive learning" is turned off in the MS speech settings. This is one time where you don't want the PC to be learning by listening to the TV. - set the time out time to be quite short. I set mine for four seconds, which is short enough not to be a problem with mistaken activations. - make sure the activation phrase has a plus in front of it. This signifies a more certain signal is needed to activate the engine. One feature request I have: It's pretty gimmicky and may be hard with the MS speech engine, but is it possible to have a command where the PC will repeat what you say? for example, you can say "Say hello to XXX" and the PC will then say "Hello XXX". Also, the voice I am now using is Callie from a company called cepstral. IT's actually very lifelike and only cost US$29. I may have already said this, sorry if I did :) and for a really off the wall request :) - Do you know of any way to access the voice engine remotely? If my housemate is watching TV, I'd like to be able to access a service from a different PC that will let me type in something for the PC to say. I figure I could have loads of fun with something like that. I'm guessing the only way would be to use a web service though. Any ideas? I'm in Mexico at the moment, but will load up the new version when I get home. [/QUOTE]