<meta http-equiv="content-type" content="text/html; charset=utf-8">Dear folks,<div><br></div><div>Thank you for your fast answer!</div><div>Well, I think that my problem was similar to what happened with Fábio Tesser. Anyways, I solved it "manually" just running linux sed command.</div>
<div><br></div><div>As I am new to TTS development, is there someone that could give me advices and references for building a BP natural voice TTS? How many sentences should I record for instance? I would like to colaborate anyhow as well.</div>
<meta http-equiv="content-type" content="text/html; charset=utf-8"><div><br></div><div>I will share some aspects, objectives and motivations, to help helping me:</div><div><br></div><div>- I just discovered Open MARY Site last week and got eager with some possibilities.</div>
<div><br></div><div>- Normally, I work as a Java Developer, coding CRUD and other boring things.</div><div><br></div><div>- Couldn't find anyone on my former University that could lead a research project on a new Open Natural BP speech solution.</div>
<div><br></div><div>- My first motivation is to create a natural voice for my wife. Yes, for my wife! She has a Retina Degeneration and can't read a complete word. Just letter by letter and with 20x lupe. Therefore, she depens strongly on voice synthesis for her studies ( Long reading of psychoanalisis texts. I thought about unit selection synthesis so far, but without a domain... how many sentences??)</div>
<div><br></div><div>- A lot of people in Brazil can't afford buying a comercial software for sight accessibility. There are nice open solutions like NVDA and Orca, but they lack good Natural voices (at least for BP). The robotic simple voices are sufficient to operate basic functions, however, for reading long text, they can cause headaches (at least for BP).</div>
<div><br></div><div>- We lead a non-profit association in our State (<a href="http://www.retinaminas.org/">http://www.retinaminas.org/</a>) that basically gives information about new researches concerning Retina Treatments and Accessibility for Sight Impaired people. It's a good idea to concentrate my development skills to create new, free and open accessibility solutions.</div>
<meta http-equiv="content-type" content="text/html; charset=utf-8"><div><br></div><div>That's it! I dream of creating a good natural sounding voice for sight accessibility, share the experiences and earn nothing for that. Could I use MARY TTS? Am I crazy? </div>
<meta http-equiv="content-type" content="text/html; charset=utf-8"><div><br></div><div>Well, I would be forever gratefull for any help on that.</div><div><br></div><div><br></div><br><div class="gmail_quote">2011/2/28 Fabio Tesser <span dir="ltr"><<a href="mailto:fabio.tesser@gmail.com">fabio.tesser@gmail.com</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">Hello Ingmar and Fábio,<br>
<br>
I would like to add my 2 cent on this:<br>
<br>
When I have tried built a new voice in a machine with an Italian locale, I remember that I got a similar error.<br>
<br>
If I remember well, the component that make use the "comma" instead the "." as my opinion was LabelPauseDeleter, but I haven’t investigated very deeply.<br>
<br>
I fixed the problem running the voice import tools from a terminal with a different locale.<br>
<br>
For example, in gnome ubuntu you can do this:<br>
<br>
$ env LANG=en_GB.UTF-8 gnome-terminal --disable-factory<br>
<br>
Best,<br><font color="#888888">
Fabio.</font><div class="im"><br>
<br>
<br>
<br>
On 02/27/2011 02:56 PM, Fábio Marinho wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Well, I think I found out the problem. It seems that the regex used in XwavesLabelfileDataSource for parsing the lab files is considering "." as decimal separator. I don't know why, maybe because of my locale, the lab files were generated using "," as decimal separator. Maybe a good bug fix should be trying to use a decimal separator from the locale of the machine executing the code.<br>
<br>
I you keep on trying here. Thank you.<br>
</blockquote>
<br>
<br>
<br></div><div><div></div><div class="h5">
On 02/28/2011 09:22 AM, Ingmar Steiner wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Dear Fábio,<br>
<br>
glad to see you found the cause of the issue so quickly!<br>
<br>
I agree that it would be slightly more elegant to support those locales<br>
that use a comma as the decimal separator (BP, German, etc.).<br>
<br>
However, I'm not convinced (just yet) that this is a clean solution. As<br>
far as I'm aware, the Xwaves lab file format does not allow commas as<br>
decimal separators, and Mary's XwavesLabelfileDataSource class is<br>
certainly not the only program that rejects files as malformed that do<br>
not adhere to the format.<br>
<br>
The question that arises is how your lab file was created. Was it some<br>
component of the Mary voicebuilding toolchain?<br>
<br>
Best wishes,<br>
<br>
-Ingmar<br>
<br>
On 26.02.2011 13:57, Fábio Marinho wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hello,<br>
<br>
I am a Java developer and personally interested in building a natural<br>
voice for Brazilian Portuguese (pt_BR).<br>
<br>
I am following all steps in "Adding New Language Support" development<br>
page of Open MARY.<br>
<br>
Everything was ok so far, but then I got stucked in<br>
PhoneUnitLabelComputer step of the ImportVoice GUI:<br>
<br>
TRACE:<br>
<br>
Computing unit labels for 15 files.<br>
From phonetic label files:<br>
/home/fmarinho/Desenvolvimento/TTS/minhavoz/lab/*.lab<br>
To unit label files:<br>
/home/fmarinho/Desenvolvimento/TTS/minhavoz/phonelab/*.lab<br>
Malformed line found outside of header:<br>
0,070000 125 _<br>
java.lang.Exception: The component PhoneUnitLabelComputer produced the<br>
following exception:<br>
at<br>
marytts.tools.voiceimport.DatabaseImportMain$8.run(DatabaseImportMain.java:294)<br>
Caused by: java.io.IOException<br>
at<br>
marytts.util.data.text.XwavesLabelfileDataSource.parseLabels(XwavesLabelfileDataSource.java:157)<br>
at<br>
marytts.util.data.text.XwavesLabelfileDataSource.<init>(XwavesLabelfileDataSource.java:71)<br>
at<br>
marytts.util.data.text.XwavesLabelfileDataSource.<init>(XwavesLabelfileDataSource.java:58)<br>
at<br>
marytts.tools.voiceimport.PhoneUnitLabelComputer.computePhoneLabel(PhoneUnitLabelComputer.java:138)<br>
at<br>
marytts.tools.voiceimport.PhoneUnitLabelComputer.compute(PhoneUnitLabelComputer.java:119)<br>
at<br>
marytts.tools.voiceimport.DatabaseImportMain$8.run(DatabaseImportMain.java:291)<br>
<br>
<br>
MY ENVIRONMENT:<br>
Ubuntu 10.04 LTS - Lucid Lynx<br>
java version "1.6.0_22"<br>
Java(TM) SE Runtime Environment (build 1.6.0_22-b04)<br>
Java HotSpot(TM) Server VM (build 17.1-b03, mixed mode)<br>
<br>
<br>
I think that maybe it could be a simple detail that I am missing. So<br>
before trying to debug the code in Eclipse, I would appreciate any help<br>
on that.<br>
<br>
Thank you in advance.<br>
<br>
<br>
<br>
<br>
_______________________________________________<br>
Mary-dev mailing list<br>
<a href="mailto:Mary-dev@dfki.de" target="_blank">Mary-dev@dfki.de</a><br>
<a href="http://www.dfki.de/mailman/cgi-bin/listinfo/mary-dev" target="_blank">http://www.dfki.de/mailman/cgi-bin/listinfo/mary-dev</a><br>
</blockquote></blockquote>
</div></div></blockquote></div><br>