similar to: agi+cepstral driving me nuts

Buenas tardes grupo. Saludos cordiales Carlos J., muchas gracias por tu orientación. Efectivamente, me había dado cuenta que la razón por la que no se aplicaba colnames era porque no tenía columnas. La cuestión es que no logro visualizar completamente/claramente en qué parte del proceso de creación del corpus se puede hacer. Sin embargo, siguiendo el ejemplo de

Hi All, We just bought a sip based PA setup here with the intention of hooking it into our existing asterisk (1.4) setup. It works as expected when I dial it's extension, but I want to have system generated speech played based on some action (using cepstral, which is already installed and working). My first thought was to DIAL to the extension, and then have cepstral play the audio. The

If you want to hear an example of the new Cepstral voices (v3.0.2) dial 700-348-3589 I have Frank and Emily available.

I just read about the partnership but was wondering what is actually going to happen? Is asterisk going to be bundled with cepstral voices for free :)? Or whats the deal?

I'm expirimenting with Cepstral via swift.agi. It loads w/ no problem, but there is a fairly long separation between words -- almost as if it is processing one word at a time, rather than stringing a phrase together. When I run cepstral with a similar script, but not through an actual call from asterisk the separation between words is short and natural. I'd really appreciate

Hello all, I am trying to get cepstral to work with Asterisk for Text to Speech purposes. I got Asterisk to recompile with cepstral and after doing *CLI> show cepstral female voices Available Cepstral voices: Name Gender Age Language Sample Rate License ========== ========= ===== ==================== ============= ============ Amy female 30

Hey everyone, I was frustrated with the existing app_cepstral/app_swift TTS modules I've found on the net, so I hacked up my own. It's been working really well for me so I thought I'd share. In developing this, I wanted to avoid: * the startup delay incurred writing TTS output to a temp file then playing it back * gluing cepstral/swift to the asterisk module in a way that fubard

Hi All, Recent discussions on app_cepstral on the list have led me to believe there's some issues with Asterisk 1.4 I set about creating a small howto for people to get cepstral, with app_swift working. Check it out: http://www.voipphreak.ca/archives/354-Asterisk-1.4-Gentoo-CepstralSwift-Howto-using-App_Swift.html Thanks, Diwelf

I'm using trixbox and the asterisk agi. I downloaded a cepstral voice and worked with it until I got the code to do what I wanted. I then registered the voice today to get rid of the 'this voice is not yet registered", stuff yet it still does that. Any ideas on how to fix this? It told me my info was valid. Thanks! -------------- next part -------------- An HTML attachment was

what's the easiest way of using cepstral voices with asterisk ? On their website, in the ssml page (http://www.cepstral.com/cgi-bin/support?page=ssml), they say "Asterisk PBX SSML can be used with Cepstral voices in Asterisk by simply embedding the markup into the input text." what input text ? To what application ? Thanks ! Julian

Does anyone have any idea on how to force cepstral to convert a number to speech ? I have noticed that sometimes it speaks the number correctly, and at others it doesn't. 1) 787 is pronounced 7-8-7 2) 123 is pronounced one-hundred and twenty-three. 1) is wrong for what i need, 2) is perfect. Is there anyway of forcing numbers to be pronounced as 2) ? I've tried looking at the ssml

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I have a license for Allison-8kHz and two concurrent port licenses that I purchased from Cepstral at the end of last year. I just got around to installing to my * 1.6.0.10 machine. I've decided that the best way for me to integrate the two would be res_cepstral, which I downloaded and installed. Everything is fine, except the register program,

Hello, I recently purchased the Cepstral 6 text-to-speech engine (swift), and am now wondering if I should have bought something else. I would like to use Cepstral text to speech like some people use the Festival() or Flite() applications. For example, when I do a "core show application flite" at the CLI, flite is available to me: localhost*CLI> core show application flite

Greetings, I've recently upgraded from Asterisk 1.2 to 1.4. I've been searching for a solution, but am also trying the easy way at the same time. I've now got David of Cepstral now speaking using app_swift from http://www.mezzo.net/asterisk/app_swift.html . The problem is, he sounds way worse than he did when it was asterisk version 1.2. I'm seriously considering either

All, Has anyone seen this before? This appears to be a Swift or app_swift bug. I'm having a difficult time finding any information or support on this. Asterisk version: Asterisk 11.6-cert4 built by asterisk @ ivrd02 on a x86_64 running Linux on 2014-08-11 13:55:25 UTC OS: Linux livrp03 2.6.32-431.11.2.el6.x86_64 #1 SMP Mon Mar 3 13:32:45 EST 2014 x86_64 x86_64 x86_64 GNU/Linux When

Was hoping someone could point me in the right direction for getting the linux version of cepstral voices working with Asterisk. The cepstral site states that the voices work with asterisk, but I haven't been able to find anything with google or in the handbook on this. Thanks in advance, Bruce

Hi all, I hope that I'm not breaking protocol too much by posting a message in this group about a problem that I'm having with an Asterisk Business Edition installation, but the reason that I'm posting here is because the problem that I'm having isn't really with the Business Edition, it is with the Cepstral text to speech product that I'm using with it, and also because

We are having some major problems with app_swift since we went live. It is regularly segfaulting. I don't know if this is my fault or not, but here's the story: Installed the cepstral voices (at the time, 4.0) on our test system (2.6.9-42.0.10.ELsmp) and later added some extra voices (now 4.2). All worked fine - we stress tested (20+ simultaneous calls). Move to live (

Hello list, I have been using Cepstral's 8KHz voices for my text-to-speech service for some time now, and have been noticing that the voice quality is really poor, doesn't matter what phrase I give it to convert. None of the other 8KHz voices I have ever used were this bad. It doesn't seem good enough system to be used in a commercial system. Is there any better quality text-to-voice

I downloaded Cepstral to my Asterisk Box. I did the install and let it install to /opt/swift. I brought down a new CVS-HEAD as of today 10/1. I added APPS+=app_cepstral.so into the Makefile in /usr/src/asterisk/apps/Makefile Like: # Obsolete things... # #APPS+=app_sql_postgres.so #APPS+=app_sql_odbc.so APPS+=app_cepstral.so # I did this piece but wasn't sure exactly what part of the