home
products
contribute
download
documentation
forum
Home
Forums
New posts
Search forums
What's new
New posts
All posts
Latest activity
Members
Registered members
Current visitors
Donate
Log in
Register
What's new
Search
Search
Search titles only
By:
New posts
Search forums
Search titles only
By:
Menu
Log in
Register
Navigation
Install the app
Install
More options
Contact us
Close Menu
Forums
MediaPortal 1
MediaPortal 1 Plugins
[Test] Plugin voice remote controle
Contact us
RSS
JavaScript is disabled. For a better experience, please enable JavaScript in your browser before proceeding.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an
alternative browser
.
Reply to thread
Message
<blockquote data-quote="kwazi" data-source="post: 50825" data-attributes="member: 20263"><p>There are several Speech Rec (SR) technologies today on the market, but I'll emphasize two differentiating factors: Speaker Dependant vs. Speaker Independent SR Engines. So to answer your question there are engines that do not require training and will understand your whole family (statistically men are understood better then women <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite8" alt=":D" title="Big Grin :D" loading="lazy" data-shortname=":D" /> and especially children under 11). The second difference is between unlimited vocabularies (aka dictation systems) vs. CFG (Context Free Grammars) or Stochastic Grammars based SR Engines. Without making it completely boring, in the first case you can say whatever you want to (but you pay very high price – high error rate and high environment requirements – silence, high quality mic etc. ); in the second case you have to define grammars, that will tell SR Engine what can be expected. I’m thinking about using Speaker independent, grammar based SR Engine and consider Sphinx as a serious candidate for the job. I used it before. Of course I would prefer to use commercial engine, such as Nuance 8.5, but I doubt anyone would spend 800$ for excellent SR Engine even if it supports 40+ languages <img src="data:image/gif;base64,R0lGODlhAQABAIAAAAAAAP///yH5BAEAAAAALAAAAAABAAEAAAIBRAA7" class="smilie smilie--sprite smilie--sprite1" alt=":)" title="Smile :)" loading="lazy" data-shortname=":)" />. What is interesting in Sphinx is that you can relatively easy add any language. Easy because the set of the sentences and commands is well defined and relatively, cause we’ll have problems with names recognition.</p><p></p><p>I’ve installed MP on my HTPC. I haven’t succeeded to configure TV Card, but this problem I’ll try to solve in appropriate topic. I am a little bit concerned about the resources MP consumes, cause Speech Rec Engine by itself might have relatively high requirements. I’ll do some experiments this weekend and keep you posted</p></blockquote><p></p>
[QUOTE="kwazi, post: 50825, member: 20263"] There are several Speech Rec (SR) technologies today on the market, but I'll emphasize two differentiating factors: Speaker Dependant vs. Speaker Independent SR Engines. So to answer your question there are engines that do not require training and will understand your whole family (statistically men are understood better then women :D and especially children under 11). The second difference is between unlimited vocabularies (aka dictation systems) vs. CFG (Context Free Grammars) or Stochastic Grammars based SR Engines. Without making it completely boring, in the first case you can say whatever you want to (but you pay very high price – high error rate and high environment requirements – silence, high quality mic etc. ); in the second case you have to define grammars, that will tell SR Engine what can be expected. I’m thinking about using Speaker independent, grammar based SR Engine and consider Sphinx as a serious candidate for the job. I used it before. Of course I would prefer to use commercial engine, such as Nuance 8.5, but I doubt anyone would spend 800$ for excellent SR Engine even if it supports 40+ languages :). What is interesting in Sphinx is that you can relatively easy add any language. Easy because the set of the sentences and commands is well defined and relatively, cause we’ll have problems with names recognition. I’ve installed MP on my HTPC. I haven’t succeeded to configure TV Card, but this problem I’ll try to solve in appropriate topic. I am a little bit concerned about the resources MP consumes, cause Speech Rec Engine by itself might have relatively high requirements. I’ll do some experiments this weekend and keep you posted [/QUOTE]
Insert quotes…
Verification
Post reply
Forums
MediaPortal 1
MediaPortal 1 Plugins
[Test] Plugin voice remote controle
Contact us
RSS
Top
Bottom