Posted: Monday Oct 29th, 2012 09:56 am |
|
21st Post |
ident
Administrator
Joined: | Wednesday Aug 9th, 2006 |
Location: | Singapore |
Posts: | 3493 |
Status: |
Offline
|
back to top
|
I am afraid it is not feasible for the protocol to provide this information
|
Posted: Monday Oct 29th, 2012 10:16 am |
|
22nd Post |
juwi_uk
Member
back to top
|
I don't understand then.
So if I have a Aud01 on the UCM then I assume the sound is sent over KA/KB to it. Couldn't at the same time also send out a phrase # (code) on the protocol that is being read out. We could then match this phrase to a list that you publish and I could then reconstruct the phrase using Myma.
ie KXnnbbpp..n where KX is command, nn is Message Code #, bb is # parabeter bytes that follow, pp..n are parameters (Count, user # etc). So a bit like the System Alarm Report (AM) command.
So say I got a KX1A0102
KX = Keypad Audio Message
1A = Code say for messages waiting
01 = 1 byte follows
02 = 2 messages
I look up Code 1A from my list of key phrases you publish, recognise the parameter is the number of messages and say "Two Messages are in your inbox".
There's surely always a solution.
Regards
Julian
|
Posted: Monday Oct 29th, 2012 10:37 am |
|
23rd Post |
ident
Administrator
Joined: | Wednesday Aug 9th, 2006 |
Location: | Singapore |
Posts: | 3493 |
Status: |
Offline
|
back to top
|
Not quite
The voice is sent by the voice wire connected to keypads and ucm/audio
The command to KA/kb is for speaker on and off only
Not the data that you need
Bus monitor will show this
|
Posted: Monday Oct 29th, 2012 10:41 am |
|
24th Post |
juwi_uk
Member
back to top
|
Ah OK,
But something somewhere is making the decision to send this audio down that cable so it could hook out to UCM/Bus to send a message couldn't it with some firmware changes? You must be doing something similar for the LED's.
Julian
|
Posted: Monday Oct 29th, 2012 12:42 pm |
|
25th Post |
ident
Administrator
Joined: | Wednesday Aug 9th, 2006 |
Location: | Singapore |
Posts: | 3493 |
Status: |
Offline
|
back to top
|
The messages led is turned on by the KL message, which tells the Keypad to turn on certain LEDs. You can certain use this to decode the new mesage led
But it does not tell how many messages are waiting
As you say that would need a neww command, which there is no cpmmand at the momem
|
Posted: Monday Oct 29th, 2012 03:07 pm |
|
26th Post |
Ingo
UCM Pi Users
Joined: | Sunday Jan 21st, 2007 |
Location: | South Africa |
Posts: | 557 |
Status: |
Offline
|
back to top
|
Very interesting thread...
Let me make a comment about installing the voice files with the ComfortClient. Cytech brings updates out every now and then, having the files 'offline' allows the user to upgrade voice files when they become available and not have to wait for a next release of ComfortClient.
Ingo
|
Posted: Monday Oct 29th, 2012 03:11 pm |
|
27th Post |
Ingo
UCM Pi Users
Joined: | Sunday Jan 21st, 2007 |
Location: | South Africa |
Posts: | 557 |
Status: |
Offline
|
back to top
|
Here is a possible 'hack' for the Remote Audio suggestion. If someone has the urge to play around with a Raspberry Pi, and some hardware, then you could tap into this audio line, read the analogue voice data, digitise it in PCM code, transfer it to the Client and then have the Client decode it on the other end...
Ingo
|
Posted: Monday Oct 29th, 2012 03:14 pm |
|
28th Post |
juwi_uk
Member
back to top
|
I don't think that would be performant enough IMHO.
I'm guessing we and Tech07 are out of luck unless Cytech can suggest a solution for this remote keyboard emulation audio.
Regards
Julian
|
Posted: Wednesday Oct 31st, 2012 02:03 am |
|
29th Post |
ident
Administrator
Joined: | Wednesday Aug 9th, 2006 |
Location: | Singapore |
Posts: | 3493 |
Status: |
Offline
|
back to top
|
Let me make a comment about installing the voice files with the ComfortClient. Cytech brings updates out every now and then, having the files 'offline' allows the user to upgrade voice files when they become available and not have to wait for a next release of ComfortClient.
The voice could be included in the ComfortCleint installation and also separately updated when a new voice version comes out
The way the sentences are announced in Comfort is, each sentence is made up of individual words in Myma or Kate
Each voice menu has a sentence and each key press may produce a new sentence
So it is quite complicated, and putting all that information on the bus may make the bus too busy and reducxe performance
I think to use the Enter Key is the simplest solution
|
Posted: Wednesday Oct 31st, 2012 05:20 am |
|
30th Post |
juwi_uk
Member
back to top
|
Ident,
I don't think this is what Ingo is saying here; ie that the audio should be streamed over the bus. What I think he's saying is how we do in ComfortClient at the moment, we construct the phrases on the fly (to a known phrase construct) via the Myma/Kate Wav files.
Surely all of the messages announced on the keypad MUST have a known construction, interspersed with some variables? All I'm asking for here is that you pass a message code and the variables on the bus (tiny data) and then I reconstruct in ComfortClient as I do with other sounds.
For example you publish the codes and phrases in XML file that define the structures; ie
<Number="1" Phrase="User {Param1} is signed in"/>
<Number="22" Phrase="User {Param1} has {Param2} messages"/>
etc
I can read and decode this file and locate the appropriate Myma WAVs for the static text.
At runtime you pass just the Code and the parameters (typically numeric I assume) as triggered.
In ComfortClient I then read out the combined phrase. The code is already doing this for System Sounds decoded from responses received on the bus so business as usual as far as I'm concerned.
This would work IMHO.
Only think we need to ensure is that additions to Myma are circulated as updates so the any new WAV's can be distributed.
You could also then use same technology in the mobile apps too.
Julian
|
Current time is 05:12 pm | Page: 1 2 |
|