Rockbox

Tasklist

FS#8589 - Allow to choose Loquendo and ViaVoice TTS

Attached to Project: Rockbox
Opened by Chelo Sacristán (csar) - Monday, 11 February 2008, 10:48 GMT
Last edited by Jonas Häggqvist (rasher) - Wednesday, 06 August 2008, 17:09 GMT
Task Type Patches
Category Build environment
Status Unconfirmed
Assigned To Jonas Häggqvist (rasher)
Operating System All players
Severity Low
Priority Normal
Reported Version Daily build (which?)
Due in Version Undecided
Due Date Undecided
Percent Complete 0%
Votes 0
Private No

Details

It would be useful to change the configure and voice.pl files in order to allow the user to choose also the Loquendo and ViaVoice TTS
This task depends upon

Comment by Chelo Sacristán (csar) - Tuesday, 12 February 2008, 07:56 GMT
Loquendo (http://www.loquendo.com) and ViaVoice (http://www.wizzardsoftware.com/) are two text-to-speech engines. They have a great quality, but, unfortunatily, they are not GPL. In order to use them to create sounds (for example the menu messages for blind people) you need the SDK and need an special licence.

'loquendo-text2file' and 'ttsynth_say' commands are referenced in the configure patch. They are to commands that send text to a sound file. Anyone can get ttsynth_say from the ViaVoice SDK. Loquendo-text2file it's a little application that i have done. If somebody need it, i can send him the source file, but rememeber that you need the Loquendo SDK in order to compile and use it.

Comment by Chelo Sacristán (csar) - Tuesday, 12 February 2008, 09:11 GMT

Another important issue is remember that ViaVoice doesn't work with utf-8. So, you must save the lang file in a compatible format with ViaVoice (ie 8859)
Comment by Steve Bavin (pondlife) - Tuesday, 12 February 2008, 09:46 GMT
Hi Chelo,

If you're using Windows, I believe that ViaVoice uses SAPI underneath their own front end, so those should work with the current code.

e.g. http://www.wizzardsoftware.com/att_desktop.php
Comment by Chelo Sacristán (csar) - Tuesday, 12 February 2008, 10:17 GMT

Hi Steve,

Yes, you are right. I forgot Windows (smile). My "habitual environment" is Linux and i use Linux SDKs to generate sound files.
ViaVoice SDK for Linux is very old, and it doesn't manage correctly utf-8 strings.

Sorry and thanks for the explanation
Comment by Chelo Sacristán (csar) - Tuesday, 12 February 2008, 12:02 GMT

Hi Steve,

Yes, you are right. I forgot Windows (smile). My "habitual environment" is Linux and i use Linux SDKs to generate sound files.
ViaVoice SDK for Linux is very old, and it doesn't manage correctly utf-8 strings.

Sorry and thanks for the explanation
Comment by Jonas Häggqvist (rasher) - Tuesday, 12 February 2008, 15:50 GMT
This looks pretty reasonable, although I think we should add charset conversion to voice.pl and introduce a VOICE_CHARSET define or something like that, which will be ignored if empty, but causes voice.pl to convert to the specified charset. I think sapi_voice.vbs might already be doing charset conversion, but I'm not sure.
Comment by Jonas Häggqvist (rasher) - Thursday, 07 August 2008, 14:39 GMT
With Jens Arnold's change yesterday, this should now be possible. I'll look into finally committing this one later today.

Loading...