Daily Reminders

[Wadenut]

I've been testing my published scripts under the Raspberry Pi version of HS3.
At this point, I'm not terribly optimistic about Daily Reminders.
The script contains support for HS Phone for retrieval of reminders over the phone line. HS3-Pi chokes on this during the compilation of the script. Unless I can devise a way to spoof hsp functions to allow the script to compile, the project probably cannot continue without developing a third function library, as I had done to support both HS2 and HS3. This would require considerable work as all three versions would have to be rebuilt. Unless there is sufficient interest in this, I'll probably not bother.

[Wadenut]

Thanks! It is installed. I will get my first reminder with a different voice tomorrow, so I will let you know.

Should work. I tested it extensively before I lost HS3.

[rprade]

version 2.4.1 is posted.
This adds the facility to handle SAPI tags within reminder texts.

@rprade
I was able to deal with the tag characters < & > without having to resort to character substitution. Double quotes however, if used, will be converted to single quotes.

Excellent suggestion, Thank-you.

Thanks! It is installed. I will get my first reminder with a different voice tomorrow, so I will let you know.

[Wadenut]

version 2.4.1 is posted.
This adds the facility to handle SAPI tags within reminder texts.

@rprade
I was able to deal with the tag characters < & > without having to resort to character substitution. Double quotes however, if used, will be converted to single quotes.

Excellent suggestion, Thank-you.

[Wadenut]

For me, the global value is not necessary. It might be a nice feature for some users. I set a default voice in the windows TTS control for all announcements. I only override the voice for announcements that may have more importance.

Sounds good to me. I do about the same thing. As I said, this would only affect VR requests and really isn't worth the effort at this point.

I'll continue testing this for a day or so, then update the doc and post the update. I thought of modifying the ASP page as well to accept the <,> and " characters. It would replace them before saving or displaying the reminders. I think however this might only add confusion for some, so will be leaving things as is.

[rprade]

OK.

Looks, or rather Sounds good.

I used the example:
This is a sapi tag test. (voice required=%name = Microsoft Anna%) (silence msec='5000'/) There should have been a delay.

- HS speaks in the system default voice (VW Paul in my case) "This is a sapi tag test" then delays five seconds and continues in Anna's voice "There should have been a delay."

- The device strings, Next Reminder, Latest Reminder, and Today's Reminders display the text without the tags.

- There are announcements embedded in the script which will still use the default voice. These are used only when reminders are requested via VR, either thru the Speaker client or by phoning in the request. I could change this if necessary by adding a Config parameter to use a specific voice for those announcements, but that would be global. Individual reminders will use the embedded SAPI tags.
Is this something you'd want?

For me, the global value is not necessary. It might be a nice feature for some users. I set a default voice in the windows TTS control for all announcements. I only override the voice for announcements that may have more importance.

[Wadenut]

OK.

Looks, or rather Sounds good.

I used the example:
This is a sapi tag test. (voice required=%name = Microsoft Anna%) (silence msec='5000'/) There should have been a delay.

- HS speaks in the system default voice (VW Paul in my case) "This is a sapi tag test" then delays five seconds and continues in Anna's voice "There should have been a delay."

- The device strings, Next Reminder, Latest Reminder, and Today's Reminders display the text without the tags.

- There are announcements embedded in the script which will still use the default voice. These are used only when reminders are requested via VR, either thru the Speaker client or by phoning in the request. I could change this if necessary by adding a Config parameter to use a specific voice for those announcements, but that would be global. Individual reminders will use the embedded SAPI tags.
Is this something you'd want?

[rprade]

OK. It seems HS has a problem with displaying text containing the tags when viewing/editing reminders in the ASP page. When I edit the data file manually however, including the unmodified SAPI tags, the reminders are spoken as expected using the alternate voice.
The best/simplest method I can think of to overcome this is to use some character substitution in the reminder texts. What I'll likely end up doing is replacing < and > with ( and ), and quotes " with %. These will be handled internally and should produce the result you're looking for.
I'll need to deal with the Device texts as well (to suppress the tags in the device strings).

For instance:
(voice required=%name = Microsoft David%)
an<code>d
(silence msec='1000'/)
</code>
Will this be acceptable?

Probably sometime next week before I can commit a large enough block of time to complete the changes.

That would work fine. No hurry!

[Wadenut]

OK. It seems HS has a problem with displaying text containing the tags when viewing/editing reminders in the ASP page. When I edit the data file manually however, including the unmodified SAPI tags, the reminders are spoken as expected using the alternate voice.
The best/simplest method I can think of to overcome this is to use some character substitution in the reminder texts. What I'll likely end up doing is replacing < and > with ( and ), and quotes " with %. These will be handled internally and should produce the result you're looking for.
I'll need to deal with the Device texts as well (to suppress the tags in the device strings).

For instance:
(voice required=%name = Microsoft David%)
an<code>d
(silence msec='1000'/)
</code>
Will this be acceptable?

Probably sometime next week before I can commit a large enough block of time to complete the changes.

[rprade]

Seems not to be that simple. Tags do need to be removed for display in device texts, which is done before the text is actually spoken, so it looks like this may take me a bit of time to accomplish. I will give the project priority but it does mean re-coding for both HS2 and HS3.

Thanks for looking. It is not a major deal. We use different voices, depending on the type of announcement and need to add a little delay to give the Sonos groups more time to build. It is working as is.

[Wadenut]

Seems not to be that simple. Tags do need to be removed for display in device texts, which is done before the text is actually spoken, so it looks like this may take me a bit of time to accomplish. I will give the project priority but it does mean re-coding for both HS2 and HS3.

[Wadenut]

Is there a way of putting SAPI tags in the spoken text. I would like to put a small delay in and change the voice on some announcements such as

 and 

Initially, I was going to say there should be no reason for this not to work, however, I've just looked at the code and it does seem that tags are removed from the text field before speaking. This may be an artifact from a previous version where redirection was included in that field. If that's the case, I should be able to do something for you. I'll need to look into it.

[rprade]

Is there a way of putting SAPI tags in the spoken text. I would like to put a small delay in and change the voice on some announcements such as

<voice required="name = Microsoft David Desktop"> and <silence msec='1000'/> 

[Wadenut]

Version 2.38 (Both HS2 and HS3)

This should be the final version.

Previously created reminder data should be fully compatible. Additional fields are created where needed when the data is first re-saved using the ASP page.

Function redirection has been broken out into a separate field making it's use simpler. Previously, redirection was embedded in the text of reminders. This method has a much cleaner look and simplifies editing of reminders. The Redirection script has been re-written so that functions are called directly rather than being processed through a "Main" routine.

Text substitution is more in line with methods already utilized within Homeseer. Includes DeviceString, DeviceValue, Date, Time and DayOfWeek substitution.

[Wadenut]

Have Redirection broken out to a separate field. Much nicer.
Also added Device String, Device Value, Date, Time and Day of week string substitution to reminder text. Very nice.

Testing some of the predesigned Redirection functions now.