search for: phonems

Hi, Andras, Thanks for the comments. Yes, I am aware of those issues. I probably should have been more accurate on my usage of terms. Actually in my project, the unit collection is a mixture of diphones and words. However seems to me, these synthesizer specific issue is irrelevant to my question about speex. As you said, i merely use speex as storage methods. All I ask for is to get the

Ok, let me first explain why 5ms matters, even they are 0's, in my particular application. I am working on a speech synthesis system. The basic idea is concatenating pre-recorded phonemes or words into longer sentences. So any missing or extra samples, even it is as short as 5~10ms, cause very noticeable discontinuities. I want to use speex to compress/decompress those pre-recorded

[At the risk of educating you about something you might already know] Natural speech in most human languages gradually changes from one phoneme to the next. Concatenating phonemes together from a fixed, prerecorded, unflexible set would give rise to abrupt changes between them (both in phoneme quality and in pitch), and thus make the resulting speech hard to understand and/or uncomfortable

are there plans to add "de-essing" as part of the speexenc procedure, to make it possible to use lower quality compression without getting those horrible computer-like "ess" sounds in the finished result? like if you say "someone said the sun is shining", there is a lot of ess sounds, and these will sound "computer-ish" at vbr qualities below 9. i know

Representative of Olav, >like if you say "someone said the sun is shining", there is a lot of >ess sounds, and these will sound "computer-ish" at vbr qualities below >9. I don't mean to be rude but what bit rate is windows media encoder encoding at and what encoder (type) are you using... Unless its low (32kbps-8kbps) it doesn't compare to speex (spx).

I'm looking at the <http://support.a-link.com/phonemate/IPU1.htm> phone because it works with Skype (from Linux), but can do SIP, too. Not necessarily asterisk related, but possibly. My networking situation might require IAX if I'm running Linux and want to use SIP, I'm not certain (Skype works fine). Putting that unknown aside for the moment, how does this phone work under

thanks for getting back to me, i have uploaded a zip file containing some sound files that demonstrates the issue. http://www.bogus.net/~olav/ess.zip this contains s.mp3 original wav file (mono) converted to top-quality mp3 (370K) s.wma windows media encoder with 19khz voice compression ( 62K) s-2.spx speexenc --vbr --quality 2 on the wav file ( 63K) s-9.spx

Hi, I think I see what you mean, though I haven't been able to listen to your wma file (not everyone has a wma decoder). The problem probably only lies in the VBR tuning for wideband which hasn't received much work yet. One way to check that is to encode in constant bit-rate and see what the results are. I'm pretty sure you'll notice the problem appears only at (CBR) quality 5 or

Hi,. We have 2 values (first formant F1, second formant F2) for a given phoneme for six languages. We want to see whether the languages are significantly different one from another for this given phoneme. We have done a manova on our data and it works well, but we doesn't allow us to see which pair of languages are different. If we have only one formant for the phoneme, we would use

Hi there, today I made the German language prompts available for download: http://www.karl.aegee.org/asterisk.nsf/HT/sound-de Be aware: Asterisk doesn't yet fully support languages other than English, there are still (smaller) issues with voicemail and date/time announcements that require a patch. Cheers, Philipp

Hi everyone! It's my first time posting to this list, and I've got a fairly technical question. I'm interested in doing phoneme extraction, and one of the first steps in the algorithm I'm planning to use is to get the LPC coefficients for an input frame. Since Speex is CELP-based, the coefficients must be generated in there somewhere. I've tried digging around in the source

Hi Jean-Marc I'm looking at the 1.0.5 source, and I'm not seeing an _spx_lpc(). There's an _spx_autocorr(), which is in lpc.c and is called near the start of the encoder function in nb_celp.c. The encoder seems to call the autocorr() function, then calls wld() to do something called Levinson-Durbin. Am I right in thinking that after the call to wld(), the st->lpc[] array

Hi list. I''m writing a program that stores a lot of Floats into MySQL database. Simplified version of the program use the following form of class. class Val < ActiveRecord::BASE end And Vals table contains one column: num double One of my data contains -Infinity for num and when I try to Val.new Val.num = <- Here goes -Inifinity Val.save! Then the program crashes:

Hi. I'm an undergraduate linguistics major and computer science minor at the University of Colorado in Boulder, and am taking a couple classes this semester which give me the opportunity to do a research project - one on introductory acoustics in the physics department, and one in the linguistics department on phonetics and phonology. I've got an idea, but I'd like to hear from anyone

Dear Sir, Could you please forward this e-mail to your engineering department. I am a ham radio person who wants to transmit from the Earth to the moon and back to the Earth by phone not by Morris code. This will be a very weak signal. I need an extra 13.8 dB of gain. Could you please help me distribute the attached paper to someone who could take this project into the next level? Please reply

Aside from the replay attacks discussed, there are some other attack vectors on the cookie_session store. I appreciate (and admire!) Jeremy''s good humor on all of this: > Planting the seed here led to quick ripening and plenty of pesticide. > Thanks for the fish, all. > > jeremy Anyway, here''s what we came up with: 1. Brute Force SHA512 can be computed _very_ fast.

** High Priority ** Hello R users My issue is both theorical and technical. I would like to run a penalised discriminant analysis with the fda() function, but I don''t know all the details of splines theory. I try on the example of the phonems from the article "Penalised Discriminant Analysis" of Hastie, Buja and Tibshirani 1994 : 5 groups and 256 variables. The 256 variables are from the discretisation of log-periodogram. Consequently, we want the scores to be smoothed by the frequency. in continuus writing, J(beta)= integra...

> >> 2. What does SPEEX_GET_LOOKAHEAD do? How to use it? >> In speexenc.c, there is following code: >> >> ... >> speex_encoder_ctl(st, SPEEX_GET_LOOKAHEAD, &lookahead); >> ... >> nb_encoded = -lookahead; >> >> Can someone explain what this means? > > The lookahead is the number of samples you need to discard at the > start.

Hello I have troubles running TextAloundMp3 under wine. TextAloudMP3 version is 1.459 Wine version is 20050725 My TextAloudMp3 directory looks like: $ ls TextAloudMp3 -rw------- 1 atom users 842 Sep 9 14:31 27bluL2.gif -rw------- 1 atom users 842 Sep 9 14:31 27bluR2.gif -rw------- 1 atom users 842 Sep 9 14:31 27blulft.gif -rw------- 1 atom users 845 Sep 9 14:31 27blurgt.gif