Active Topics

 



Notices


Reply
Thread Tools
benny1967's Avatar
Posts: 3,790 | Thanked: 5,718 times | Joined on Mar 2006 @ Vienna, Austria
#51
Just something to play with (maybe some of you knew, I didn't until now):

espeak - and therefore espeakcaller - supports SSML (Speech Synthesis Markup Language) tags in the text if the "-m" switch is added to the command line (tab "Misc." in espeakcaller).

I now use SSML to add emphasis to certain words (the "<emphasis>"-element) or change the language within a string. (Say the English word "Sir" within a German sentence using "<voice xml:lang="en">").

The espeak documentation says:
-m
Indicates that the text contains SSML (Speech Synthesis Markup Language) tags or other XML tags. Those SSML tags which are supported are interpreted. Other tags, including HTML, are ignored, except that some HTML tags such as <hr> <h2> and <li> ensure a break in the speech.
So you might also use some simple HTML like <hr> for breaks, if this is useful.
 

The Following User Says Thank You to benny1967 For This Useful Post:
benny1967's Avatar
Posts: 3,790 | Thanked: 5,718 times | Joined on Mar 2006 @ Vienna, Austria
#52
there's one thing i've always hated in the S60 speech synthesizer when it pronounced names:

i'm living in austria and most of the people in my contacts database have names that work well with a german pronunciation. however, some people have french, turkish or croatian names that sound awful when the speech synthesizer is set to german.

it would be nice if some fure version (2.0, 3.0,... ) of espeakcaller could allow settings per number. like: you define all settings as you do now, but then you can add certain numbers to the espeakcaller-database and have distinct settings for each of these numbers. this would allow me to set the language differently for some callers, but it would also have the nice side effect that you could have other voices, different texts or maybe even no text at all for some of your contacts.
 
Posts: 30 | Thanked: 21 times | Joined on Jan 2010
#53
Originally Posted by Arto View Post
This option is still under development. Maybe in next version




There are basicly same options for this under "Unknown caller"-tab.

Just change "Call from %number%" and "Call from blocked number." to "Unknown caller calls" or whatever you like


Arto
sorry i ment to say the private number text input field

Edit. now i get it, by blocked number you mean private/witheld number

Last edited by Kyaner; 2010-03-27 at 16:57.
 
benny1967's Avatar
Posts: 3,790 | Thanked: 5,718 times | Joined on Mar 2006 @ Vienna, Austria
#54
It's the most fantastic piece of software I've ever heard.
 
Posts: 10 | Thanked: 4 times | Joined on Mar 2010
#55
This is one of the functions i have been missing up until now. I was in the process of learning a bit of python and dbus-stuff to get the caller-id to 'print' but you beat me to it! (and saved me the hassle of spending a week on learning it all :-) )

In short: Thanks for this mighty fine piece of software!
 
Posts: 30 | Thanked: 21 times | Joined on Jan 2010
#56
And now it's broken
It just stopped working for me, i uninstalled/rebooted/reinstalled it but still nothing, it tried to speak at one of my tries but it was garbled...
Still the synthesizer works preoperly through xterminal....

Any ideas ?

Thanks in advance
 
benny1967's Avatar
Posts: 3,790 | Thanked: 5,718 times | Joined on Mar 2006 @ Vienna, Austria
#57
same problem now. dont have time to investigate ATM, but it doesn't make a sound when the call comes in. test-button in the GUI works fine, though.
 
Posts: 203 | Thanked: 445 times | Joined on Mar 2010
#58
this is cool stuff!

it works for me so long as there's no umlauts in the text (either directly or in the callee's name). the test from the ui works with umlauts, though.
 
Posts: 48 | Thanked: 5 times | Joined on Mar 2010
#59
Hi Arto !

I had found an important "bug" i would like ot report :

Your great application is dependent on : maemo-python-device-env 1.0.0-3.

This is not an issue UNTIL an other application installed which is ALSO dependent on the same package. As soon as it happens, both application fail to operate.

If you wish to test this, simple installe e.g: smscounter widget, and here you go.

May I ask to check how you can compile or such WITHOUT this package ? I am not a programmel, but there are tons od python applictaion does not using this, so surely possible.

Many Thanks for your soonest answer,

Andrew
 
Posts: 12 | Thanked: 58 times | Joined on Mar 2010 @ Finland
#60
I just uploaded version 0.4 to extras-devel and garage.
Changelog:
Code:
UI changes:
- Added: Option to repeat saying caller name.
  Note: This is done using SSML markup so -m flag is enabled by default

Daemon changes:
- Fixed: "eSpeakCaller still has no clue about caller name"-bug
  (Phonebook number containing spaces should work)

- Fixed: "Me speaks no umlauts"-bug
 (Umlauts & other non-ascii character should now work (finally?))
Originally Posted by coolice View Post
Your great application is dependent on : maemo-python-device-env 1.0.0-3.
Removed depency as worked fine for me without it.

Originally Posted by pillar View Post
@Arto: Don't forget the important feature to cancel the saying of the name IF the phone call is answered beforehand.!
Try new version. Tell if this works.

Originally Posted by benny1967 View Post
one last thing would be to have more than one variable for parts of the name. e.g. %first% for first name, %last% for last name, %nick% for nickname, %name% for full name...
This is still under development

Originally Posted by benny1967 View Post
it would be nice if some fure version (2.0, 3.0,... ) of espeakcaller could allow settings per number.
Maybe in future. Might take a while so I don't make any promises

Arto
 

The Following 9 Users Say Thank You to Arto For This Useful Post:
Reply

Tags
espeakcaller


 
Forum Jump


All times are GMT. The time now is 09:54.