Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2004-04-10

00:00:09midkhey all
00:01:16 Join eixei [0] (~amb@pD9E4C3BA.dip.t-dialin.net)
00:01:22eixeiHi
00:01:26eixeiAnyone awake?
00:01:27LinusNyo
00:01:31*midk is
00:01:34eixeigood :-)
00:01:51eixeiI have a non-rockbox related question, but it would be great if you could help me
00:01:59eixeiIs it possible to boot from an archos jukebox?
00:02:15LinusNit all depends on the PC BIOS
00:02:20midksome recent bios releases let you boot from a usb device
00:02:32eixeiHmm, so on them it would be possible, right?
00:02:50LinusNand you won't be able to boot from a Studio/Player, only from a Recorder
00:03:04eixeiOk, I got a recorder :-)
00:03:14eixeiGreat! Thanks a lot. And btw, rockbox is great!
00:03:22LinusNthx
00:14:52 Quit top_bloke (Read error: 60 (Operation timed out))
00:16:26midki'm going to make a gmini type charging screen :D
00:21:53 Join diddystar5 [0] (~lee@IC83.library.oregonstate.edu)
00:32:48 Join amb_ [0] (~amb@pD9E4C3BA.dip.t-dialin.net)
00:32:55 Quit eixei (Read error: 110 (Connection timed out))
00:33:00 Quit amb_ (Client Quit)
00:35:44midkhey Linus, i hope you don
00:35:58midkhey Linus, i hope you don't mid that i shortened your name to "Linus N. Feltzing" in my credits roll?
00:36:05midkyour full name doesn't fit
00:36:20LinusNwhich credits roll?
00:36:25diddystar5it dosent fit in the normal credits either....
00:36:27midkclock credits roll
00:37:01amiconnmidk: I'm in the process of speeding up grayscale by another ~14%...
00:37:18midkamiconn: great :)
00:37:42midki'm not using it in this release but i'll toy with it when i finish this revision
00:38:27diddystar5amiconn - LinusN, i got my archos green light to come on, but then there was a short :(
00:38:38midki was thinking of using it for a transition
00:38:42midkIE fade between black?
00:38:45diddystar5so it does still show a small sign of life
00:38:49LinusNsaw that in the logs, at least a sign of life...
00:39:01midkso Linus, no prob?
00:39:08LinusNnope
00:39:53amiconnLinusN: I measured power consumption (with my recorder) in various states including grayscale display.
00:41:22amiconnThis proved my theory: the power consumption of the display does not increase measurably when doing grayscale, it depends on CPU load.
00:41:49LinusNnice
00:42:37amiconnMeasured values: Idle - 97.3 mA, Grayscale display (only): 103.9 mA, Playing music (with peak meter in WPS): ~106 mA (varies a bit)
00:42:57amiconnMusic and grayscale: ~106 mA
00:43:42midkso grayscale takes up about as much as the peak meter?
00:43:44midk:)
00:44:13amiconnBtw: While measuring, I managed to get my box to RLD _once_! Couldn't repeat that.
00:44:29amiconnNever got RLD before (I have an IBM hd)
00:46:02amiconnSo possibly RLD is caused by flakey power alone, nothing that Rockbox could avoid...
00:47:53LinusNcould be
00:48:10amiconnI also noticed that my batteries start to degrade (still the original ones): There was a tiny bit of white powder near the plus contact of one of them.
01:00
01:08:57 Join AciD [0] (~acid@longchamp44-1-82-67-133-87.fbx.proxad.net)
01:35:19 Join Ka__ [0] (~tkirk@pcp261336pcs.howard01.md.comcast.net)
01:35:19 Quit Ka_ (Read error: 104 (Connection reset by peer))
01:40:46 Quit funkymonkey (" I love my HydraIRC -> http://www.hydrairc.com <-")
01:43:36 Join FBZ [0] (~SSFBZ@dhcp065-024-092-041.columbus.rr.com)
01:43:39FBZanybody there
01:43:47 Quit AciD (Read error: 54 (Connection reset by peer))
01:44:00FBZ...maybe...maybe not?
01:46:10LinusNyo
01:46:14FBZhoy
01:46:16FBZfirst off
01:46:31FBZI bought an Archos Jukebox FM Recorder 20 today
01:46:40FBZI instantly downloaded and placed the Rockbox OS
01:46:47FBZbecause it seemed like a good idea
01:46:55FBZanyhoo, thank you guys for Rockbox, this thing is awesome
01:47:09FBZMy question is, how do I make my MP3 player keylock
01:47:12LinusNglad you like it
01:47:20FBZI was unable to find info anywhere on how to do that
01:47:32***Saving seen data "./dancer.seen"
01:47:33LinusNin the WPS, press F1+Down
01:47:35FBZI was wondering if there was a setting inside hidden in the GUI
01:47:51FBZthank you so much
01:47:56FBZthat's all I needed
01:48:01FBZThis player was $270
01:48:05FBZI got it for $67
01:48:09FBZ75% off
01:48:13midkwow
01:48:14midkcool!
01:48:19FBZwarehouse was trying to get rid of them
01:48:21FBZi work at mediaplay
01:48:23LinusNnice
01:48:24FBZand we got 2 in
01:48:28FBZa customer bought one
01:48:36FBZi hid the other and bought it today (payday)
01:48:40FBZthis thing is so awesome
01:48:49FBZsomewhat bulky but that's alright, I won't use it for workout
01:48:55FBZonly when I'm driving, or walking to campus
01:48:58FBZand your OS is solid
01:50:12LinusNFBZ: page 33 in the manual :-)
01:50:42FBZlol i knew it was awesome
01:50:46FBZi knew it was somewhere i mean
01:51:06FBZI'm so glad I didn't buy the Creative Jukebox $300 30gb
01:51:25diddystar5heh yeah
01:53:00FBZwell thanks for the tip
01:53:10FBZand keep up the good work, rockbox people
01:53:14FBZhave a good weekend
01:53:28 Quit FBZ (Remote closed the connection)
02:00
02:00:47midkLinus:
02:00:49midkhttps://sourceforge.net/tracker/index.php?func=detail&aid=932605&group_id=44306&atid=439120
02:02:22midkhopefully you can merge it if there isn't anything wrong with it
02:02:28midkit fixes a few major bugs
02:03:22 Join o0Katz [0] (~poop@bgp928042bgs.brghtn01.mi.comcast.net)
02:03:28o0Katzhey guys
02:03:34midkhey
02:03:45o0Katzcould someone help me out with setting up my voice?
02:04:03o0Katzwhere do i put the voice file?
02:04:09midkin .rockbox/langs
02:04:32o0Katzthanks
02:04:39LinusNmidk: why did you move the settings file?
02:04:55midklinus: i thought it just didn't fit in the rocks folder
02:05:01midki always saw it when i went thre
02:05:11midki thought it sounded better in .rockbox where people don't browse as often
02:05:21midkand it fits in among the other settings files
02:05:22LinusNwell, it belongs to clock.rock
02:05:37midkyeah but it's at the top of the list
02:05:41midkyou can change it
02:06:15LinusNi didn't know that the settings shows when you browse the plugins
02:06:19midkthanks for asking
02:06:25midkwell if you go right into .rockbox/rocks
02:06:27o0Katzhow can i lower the voice volume?
02:06:29LinusNit should only display the .rock files
02:06:41midko0Katz: can't
02:06:45o0Katzrats
02:06:51o0Katzoh well, this is an awesome feature anyway
02:06:52midkit uses the global sound setting volume
02:06:55o0Katzoh
02:06:55o0Katzok
02:06:58o0Katzwell its not too loud
02:07:06midkLinusN: i just meant that if you were browsing instead of using the menu
02:07:10midkit fits in with the other settings
02:07:11o0Katzreally neat nonetheless
02:07:31LinusNmidk: i don't agree that it fits in
02:07:38midkLinus: go ahead and change it
02:07:43LinusNi will
02:07:46midkbut make sure you warn them to delete their old ones first
02:07:49midkor it may get screwy
02:08:16LinusNyour plugin should be able to detect that
02:08:24midkit's not built to...
02:08:42LinusNthen make it so
02:08:42midkdoes rockbox detect if i open .playlist_control and input random values?
02:09:04LinusNi think so
02:09:19midkoh :D
02:09:22LinusNbut that is not a valid case
02:09:23midkwell
02:09:24 Part o0Katz
02:09:51midkok how about changing it to .rockbox/rocks/.clocksettings :)
02:09:56LinusNit should be easy for you to have a version number in the settings file
02:10:14midkbut
02:10:20midkthe old one didn't have a version number
02:10:26midkso either way it's gonna have to be deleted
02:10:32LinusNno
02:10:34midkor we can make them load a different update that adds one
02:10:43midkor we can just tell them to delete it
02:11:08LinusNyour file parser should of course be able to see if there is no version number
02:11:19amiconnmidk: Just uploaded my improved grayscale code, will now try your new clock.
02:11:19LinusNand ignore the file
02:11:45midkwell linus, it should work ok with an old settings file
02:11:54midkit won't have the default settings
02:12:01midkthat's all
02:14:42midkamiconn: ok :) i'll have a look as well
02:15:12amiconnDrawing time for my demo is now just under 1 second.
02:15:35midkcool
02:15:39midk*likes the demo
02:15:46midk*stares at it for long periods of time
02:16:14amiconnWell, the demo itself is not that interesting...
02:17:05 Join mecraw_ [0] (~mecraw@c-24-9-226-58.client.comcast.net)
02:18:35midkoh Linus I have an idea
02:18:48 Part mecraw_
02:19:02midkwhat if i just add in some code that measures the size of the file and if it's not large enough, it ignores it?
02:19:06midkthat ok?
02:20:30LinusNthat's one way of doing it
02:20:39midkis that an okay way of doing it?
02:20:55midkor do you want it done a different way?
02:21:48LinusNit's ok with me, although the version number is surely a better way
02:21:53amiconnmidk: Gave your new clock a try - looks really nice.
02:21:59midkamiconn: thanks
02:22:11midkLinusN: but the original *has* no version number
02:22:15midki don't see what you're getting at
02:22:49amiconnJust found two small bugs/quirks: (1) The logo chooser says "On to exit" but it does not exit with o
02:23:02amiconn..with "On" but with "F2"
02:23:18LinusNmidk: so if the file has no version number, your file parser should ignore the file
02:23:54amiconn(2) You have an option to choose 12/24h display, but no option to choose European vs. American date format.
02:24:18midkamiconn: or asian format :D
02:24:43amiconnAmerican: 4/10 vs. European 10.4.
02:24:52midkyeah
02:25:02midkand asian is DD/MM/YYYY i believe
02:25:23midkhow would it know if it were a version number, then, linus?
02:25:39amiconn..which is the same as European, except for the different delimiter.
02:25:58midkit could confuse a true bool with a "Version 1" settings file
02:27:05LinusNmidk: you have a binary file format???
02:27:27midki believe so
02:27:50LinusNbelieve?
02:28:04midkthink, believe..
02:28:28LinusNoh, you dump the struct to a file...
02:28:44midkyers
02:28:45midkyes*
02:28:51midkhey hey
02:28:57midkif(sizeof(fd) == sizeof(struct saved_settings))
02:29:01midkthis works right?
02:29:14LinusNi suggest you have a version number first in the struct, and also check the file size
02:30:02midkbut if it read a true bool as a version one how would that do anything
02:30:51LinusNwhy would it?
02:31:03midkwhy would it not?
02:31:15midkif it looks for the first thing in a struct
02:31:36LinusNyou defined the struct
02:32:10midkehm, the filesize check doesn't seem to work
02:32:23LinusNif you introduce a version number in your struct, you can use that for future file incompatibility
02:32:55LinusNand you use the file size to solve the first problem, where there is no version number in the current files
02:33:11LinusNsizeof(fd) is always 4
02:33:11 Quit scott666_ ("i'll be back...eventually...")
02:34:09midkfor the old settings
02:34:10midknot the new one
02:35:01LinusNsizeof(fd) is always 4, since sizeof() tells you the size of the variable, not the file
02:35:41midkwell i can't do this
02:41:24midkuploaded a new file removing tab indents
02:41:58amiconnLinusN: two days ago, you said something about logic analyzer results...
02:42:41LinusNyeah i know
02:43:11LinusNthat'll have to wait until i get home
02:43:17LinusNnext week
02:43:44amiconnOk then. Just thought you did them already since you mentioned..
02:44:43amiconnAlready tried my recording patch (without having an LA at hand)?
02:44:55LinusNnope
02:45:35diddystar5see you
02:45:37 Quit diddystar5 ("Leaving")
02:47:33midklinus: how about this?
02:47:34midkif(sizeof(rb->open(default_filename, O_RDONLY)) == sizeof(struct saved_settings))
02:48:51 Join yozac [0] (~chatzilla@h00062506b8d9.ne.client2.attbi.com)
02:48:58amiconnmidk: no way with soizeof(), this is for variables only and is evaluted at compile time.
02:49:53yozachas there been any resolution for the red light of death problem? If not is there a place where all the info is collected so I can easily read about it?
02:49:54LinusNmidk: int filesize(int fd)
02:50:21yozaci've been experiencing it jogging recently
02:50:29midkwell does fd contain the size of the struct?
02:50:35LinusNyozac: i did some changes that may have helped the situation, but so far it seems like a hardware problem
02:51:01amiconnmidk: no, fd is the file handle
02:51:06LinusNmidk: no, but filesize(fd) returns the size if the file
02:51:14yozacLinusN: thats too bad. in the little reading I did, it seemed people were saying that the archos firmware doesn't have that problem. That turns out not to be true?
02:51:21midkLinusN: then that would work, no?
02:51:53yozacLinusN: I'm running a daily build from a few days ago, so it probably has your changes, no?
02:52:41LinusNyes it has
02:52:52LinusNand yes, the original firmware has it too
02:53:06midk"has it" the rlod?
02:53:08yozacLinusN, hmmm . . . thats too bad . . . oh well
02:53:08 Quit yozac (Client Quit)
02:53:17midki thought it was a rockbox-only problem
02:54:54midkokay
02:54:57midkthis then would do it
02:54:58midk if(rb->filesize(fd) == sizeof(struct saved_settings))
02:55:37LinusNyup
02:55:52midkand then the version
02:56:06midkfor this i could just put int version = 2; right in the struct
02:56:09midkright?
02:56:25LinusNyes, first in the struct
02:59:40midkshould i just add to the existing if case for the settings check or do a "sub-if", ie, if the settings file is the right size it goes into the next if, if the version is right?
03:00
03:00:58LinusNcheck both size and version
03:01:07midkright..
03:01:10midkbut i was just thinking
03:01:40midkwell i don't see how you can check that, if we look at settings.version that's the clock's hardcoded version right?
03:01:48midkversion 2 in this case
03:02:55LinusNyes
03:03:18midkso how will we check the actual file's version?
03:03:34LinusNby reading it of course
03:03:40midkwell yeah
03:03:44midkhow would we read just that part?
03:04:01midki see how to read the whole thing, but not just to compare one line to the clock's version
03:05:27LinusNread the whole thing
03:05:45LinusNafter checking that the file size matches
03:05:55LinusNor read only the first 4 bytes
03:06:20midkmind if i paste 6 lines?
03:07:35LinusNgo ahead
03:07:41midk if(fd >= 0)
03:07:41midk {
03:07:41midk if(rb->filesize(fd) == sizeof(struct saved_settings))
03:07:41DBUGEnqueued KICK midk
03:07:41midk {
03:07:41midk rb->lcd_puts(0, 7, "Loading...");
03:07:41***Alert Mode level 1
03:07:41midk rb->read(fd, &settings, sizeof(struct saved_settings));
03:07:43midk rb->close(fd);
03:07:45midk rb->lcd_puts(0, 7, "Loaded settings");
03:07:47midk }
03:07:49midk else
03:07:51midk rb->lcd_puts(0, 7, "Old settings file");
03:07:53midk }
03:07:55midkthis looks okay right?
03:08:28LinusNyup
03:08:35midkso we read the file
03:08:51midkand apparently store its contents in saved_settings
03:08:51LinusNand then check the version
03:08:56midkrght
03:09:02midkhow do we read a single line?
03:09:06midki see we load it
03:09:14amiconncu later..
03:09:23midkbye amiconn
03:09:32 Part amiconn
03:10:17LinusNjust read the entire struct, just like your code above, then check the version number
03:10:37midkbut the clock has a hardcoded version number
03:10:43midkyou mean check settings.version?
03:11:23midkOH
03:11:23LinusNof course
03:11:25midkI KNOW
03:11:27midkwait
03:11:32midkwe'll have a seperate
03:11:36midkint version = 2;
03:11:41midkand in the struct
03:11:43midkint versionl
03:11:46midkversion; *
03:12:00midkand we can compare the version with the settings.version
03:12:01midkright?
03:12:25LinusN#define CURRENT_VERSION 2
03:12:36midkwell that works yeah :)
03:12:41LinusNif(settings.version != CURRENT_VERSION)
03:12:42midki'm a genius
03:14:36midkthen we end up with this
03:14:37midk if(fd >= 0)
03:14:37midk {
03:14:37***Alert Mode level 2
03:14:37midk if(rb->filesize(fd) == sizeof(struct saved_settings))
03:14:37***Alert Mode level 3
03:14:37midk {
03:14:37***Alert Mode level 4
03:14:37midk if(settings.version == CURRENT_VERSION)
03:14:39midk {
03:14:41midk rb->lcd_puts(0, 7, "Loading...");
03:14:43midk rb->read(fd, &settings, sizeof(struct saved_settings));
03:14:45midk rb->close(fd);
03:14:47midk rb->lcd_puts(0, 7, "Loaded settings");
03:14:49midk }
03:14:51midk else
03:14:53midk rb->lcd_puts(0, 7, "Old settings file");
03:14:55midk }
03:14:57midk else
03:14:59midk rb->lcd_puts(0, 7, "Old settings file");
03:15:01midk }
03:15:03midk else
03:15:05midk rb->lcd_puts(0, 7, "Can't find settings");
03:15:07midkeek large sorry
03:16:02LinusNlooks ok
03:17:22midkok
03:17:28midkanything else need to be done or is it ok after that?
03:21:15LinusNapart from amiconn's findings, no
03:21:24midkoh yeah
03:21:50midki'll have to toss a feature or implement scrolling...
03:23:04LinusNgotta sleep, nite
03:23:11midknite
03:23:13 Part LinusN
03:24:38***Alert Mode OFF
03:43:35 Join diddystar5 [0] (jirc@IC98.library.orst.edu)
03:43:58 Join AciD [0] (~acid@longchamp44-1-82-67-133-87.fbx.proxad.net)
03:47:33***Saving seen data "./dancer.seen"
04:00
04:15:45 Quit diddystar5 ("Leaving")
04:16:10 Join diddystar5 [0] (lee@IC98.library.orst.edu)
04:16:48 Nick diddystar5 is now known as dstar (lee@IC98.library.orst.edu)
04:43:19dstarbye bye
04:43:31 Quit dstar ("Leaving")
05:00
05:00:46 Quit _aLF (truong.freenode.net irc.freenode.net)
05:00:46NSplittruong.freenode.net irc.freenode.net
05:00:46 Quit mbr (truong.freenode.net irc.freenode.net)
05:00:59NHealtruong.freenode.net irc.freenode.net
05:00:59NJoin_aLF [0] (~Alexandre@mutualite-3-82-67-66-128.fbx.proxad.net)
05:00:59NJoinmbr [0] (~mb@stz-softwaretechnik.com)
05:25:49 Quit AciD (Read error: 104 (Connection reset by peer))
05:26:25 Join scott666_ [0] (scott666@c-24-245-58-245.mn.client2.attbi.com)
05:26:42midkscotteh
05:47:34***Saving seen data "./dancer.seen"
07:00
07:02:39midkrebooting brb
07:04:58 Quit midk (Read error: 104 (Connection reset by peer))
07:08:41 Join midk [0] (mk@AC846D8F.ipt.aol.com)
07:47:38***Saving seen data "./dancer.seen"
07:57:48 Join methangas [0] (methangas@0x50a46197.virnxx10.adsl-dhcp.tele.dk)
08:00
08:02:18midkmethane GAS! RUN!
08:02:32methangasmorning
08:02:39midk:D hey
08:02:40midknite
08:02:41midk:)
08:03:14methangasi played starcraft 'till 03:00 last night
08:03:28methangasand now the clock is 08:00
08:03:33midkwow
08:03:38methangasand i can't sleep any longer
08:04:16methangasmeh
08:04:23CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
08:04:23*methangas goes back to being foggy
08:18:03midknice job methane
08:18:05midkgas.
08:20:59 Quit midk ()
09:00
09:47:40***Saving seen data "./dancer.seen"
09:52:31 Join pfavr [0] (pfavr@t3o902p465.telia.com)
11:00
11:01:33 Join Mine78-Tensore [0] (Mine78@host212-3.pool8249.interbusiness.it)
11:01:41Mine78-Tensorehi guys
11:01:59Mine78-TensoreI had something to post about voice files
11:02:10Mine78-TensoreI have a spelling database in italian
11:02:19Mine78-Tensorei suppose it can be used with rockbox
11:03:18Mine78-Tensorethe file is .WDB
11:03:35Mine78-Tensorethe applcation is edit2speech
11:03:46Mine78-Tensoreis there anyone ?
11:13:45 Quit pfavr (Read error: 110 (Connection timed out))
11:17:21_aLFMine78-Tensore> make a patch if the files aren't too big
11:18:45Mine78-TensoreI have the .WDB file but i don't know how to manage it
11:18:50Mine78-Tensorecan I send it to you ?
11:21:22_aLFI can't find a site to download edit2speech
11:22:20Mine78-TensoreI have all in one file
11:22:23Mine78-Tensoreit's 2.5 Mb
11:22:47_aLFyes, send it
11:23:01Mine78-TensoreI am sending
11:23:30Mine78-TensoreThe program was originally inb english (in fact it can pronounce some english word correctly) but had been translated in italian to be used on local radios
11:24:04Mine78-TensoreI hope it could be useful for the voice translation project
11:25:48Mine78-Tensoredo you tink it could ?
11:28:12_aLFnot easy to use a script because it's windowed application
11:29:10Mine78-TensoreSo it will not be of any use ?
11:30:38_aLFit's hard because we need for each translation to copy/paste the text, start recording, start speech, end recording
11:31:05_aLFwe=the person who make the voicefont
11:31:51Mine78-Tensoreanyway if it can be used I will be happy...
11:33:34_aLFI put it here : http://a.flament.free.fr/data/Voice.rar
11:35:51Mine78-Tensoreso anyone can explore it !
11:36:27_aLFexcept if you want ?
11:36:44Mine78-TensoreOf course I want !
11:36:57Mine78-TensoreI will like italian language on my rockbox ;-)
11:37:27_aLFI was looking for http://freshmeat.net/browse/124/?orderby=&offset=20
11:37:43_aLFif there is something interesting
11:38:22Mine78-Tensorewow many cool softwares !
11:39:37Mine78-Tensoremaybe there some info about the wdb file format:
11:39:37Mine78-Tensorehttp://www.asel.udel.edu/speech/Spch_proc/libcwav_functs/libcwav_funct_list.html
11:40:13methangasspeaking of
11:40:52methangasi recently downloaded the english voice pack, but i have somewhat failed to find any documentation on how to get the voicepack working
11:41:01methangascould someone give me a hint or anything?
11:42:05Mine78-Tensoreyou must put the voice files in the lang directory
11:42:20Mine78-Tensoreand then choose one of the 2 languages supported until now
11:42:24Mine78-Tensoret works
11:42:49methangasah
11:43:02methangasi thought i had to choose the voice file directly
11:43:03methangasthanks
11:43:14Mine78-Tensorenothing I have justr figured out it
11:43:46Mine78-Tensorehttp://www.avisynth.org/ doesn't works ! Does anyone have the file to send me ?
11:47:44***Saving seen data "./dancer.seen"
11:51:46_aLFhttp://www.cstr.ed.ac.uk/projects/festival & http://tcts.fpms.ac.be/synthesis/mbrola.html seem to support many language (see 2nd link)
11:57:33 Join [IDC]Dragon [0] (~idc-drago@pD9FF8C54.dip.t-dialin.net)
12:00
12:09:21Mine78-TensoreEMBOLA ! COOOL !
12:15:03_aLFfound this url for different languages : http://www.scansoft.com/realspeak/demo/
12:20:37 Join amiconn [0] (~jens@pD9E7F0C2.dip.t-dialin.net)
12:21:33amiconnHi Jörg!
12:24:52amiconnI have a question concerning to voice UI.
12:24:59amiconn*the
12:28:48Mine78-TensoreI have to go
12:28:52Mine78-Tensorebye and thanks
12:29:03 Quit Mine78-Tensore ()
12:29:39[IDC]DragonI'll be avail soon, sorry
12:38:51[IDC]DragonOK now
12:40:01amiconnWell, your deustch.voice does not sound very pleasant, so I consider ordering TextAuldoud MP3 along with AT&T natural voices and the 2 german voices. Would it be worth it?
12:40:10amiconn*deutsch.voice
12:40:50amiconnGrr, *TextAloud
12:41:08[IDC]Dragonwhich voices?
12:41:32amiconnThere are only 2 german voices: "Klara" and "Reiner".
12:42:40[IDC]DragonYou got those?
12:43:10amiconnNo, I consider buying those (see above) and would like to know if it's worth it.
12:43:49[IDC]DragonCan sound very nice, have you tried the web demo?
12:44:23amiconnTextAloud MP3, AT&T natural voices and the german voices would sum up to ~€ 110,- including shipping (at ShareIT)
12:44:41amiconnYes, I tried the web demo, sounds really nice.
12:44:51[IDC]Dragonso you also get the engish voices?
12:45:09amiconnYup, these are included with the engine.
12:45:14[IDC]Dragonnot bad.
12:45:54[IDC]DragonYo don't really need TextAloud, I don't know if just the voices are avail.
12:46:01[IDC]DragonYou
12:46:56amiconnThe prices are a bit strange for me, though: TextAloud costs € 21,-, Natural voices (engine and english voices) also € 21,- (incl. 30 % rebate).
12:47:17amiconnBut: each additional voice costs € 30,-
12:47:30[IDC]Dragonhmm
12:48:09amiconnNo, you have to buy TextAloud as well as the natural voices are not sold without it.
12:48:46[IDC]Dragoncan you buy add-on voices later?
12:49:13[IDC]DragonHow do they prevent you from just buying the voices then?
12:50:52amiconnYes, you can. You have to proof your purchase of one of the TextAuloud products. This can be verified by NextUp.com. since they only sell by themselves.
13:00
13:00:12[IDC]DragonNumders are a real problem.
13:00:17[IDC]DragonNumbers
13:00:57amiconnNot really if it _is_ possible to build the numer speaking algorithm into the voice file.
13:01:02amiconn*number
13:01:15[IDC]DragonI certainly don't want to blow up the core code with localized implementations.
13:01:38[IDC]Dragonhuh, really challenging.
13:01:51[IDC]DragonI prefer to ignore that problem for now.
13:01:56amiconnI'm not saying to build that into the core. but into the *voice file*
13:02:02[IDC]DragonI know.
13:02:46[IDC]DragonThat would better be position independent code, something the Gnu compiler/linker is not good in.
13:03:08[IDC]DragonI experimented with such for the plugins
13:03:38amiconnSo you would have a function in within the voice file, e.g. speak_number(). You would call this from the main code with e.g. speak_number(21), and it would say "twenty one" in english. but "einundzwanzig" in german.
13:03:52[IDC]Dragonthe code is not position independent, but relocatable by offsetting a table within.
13:04:14amiconnThe plugin code is also relocated on load, right?
13:04:24[IDC]DragonThe function would probably not speak, but generate a set of IDs.
13:04:42[IDC]DragonNo plugins are linked to fixed addresses.
13:04:52amiconnAhh.
13:05:06[IDC]DragonThats' why 2MB and 8MB plugins are incompatible.
13:05:55[IDC]DragonI had some discussions with Björn, but he/we were too lazy for a more complex implementation.
13:06:09amiconnSo would it be possible that some code in the main core memcpy()s the function into the plugin space (which would be constan7t so code can be linked to a fixed address)
13:06:44[IDC]DragonYes, we could abuse the plugin space.
13:07:07[IDC]DragonBut after each plugin start this would need to be reloaded.
13:07:31[IDC]DragonThe number speaking could actually be a plugin
13:07:32amiconnAm I correct: the current UI does not allow running plugins in parallel to the UI?
13:07:59[IDC]DragonA plugin runs in the foreground
13:08:33amiconnThat's why I did not suggest to let it be a plugin, but to use the plugin space.
13:08:34[IDC]DragonI once had a kind of TSR plugin
13:08:51[IDC]Dragonit kept running it's ISRs
13:09:09[IDC]Dragononce you loaded another, it crashes
13:09:48amiconnJust thought about that. The core code could copy the function over before any number speaking, just to make sure.
13:10:22[IDC]DragonI think having the "number speaker" as a plugin is not a bad idea.
13:10:39[IDC]DragonWith some extension of the plugin handling.
13:10:59amiconnMaybe, since it is already possible to hand over a parameter to a plugin.
13:11:14[IDC]DragonSo, the loader can check if the plugin has previously been started, to prevent loading it from disk again.
13:12:27amiconnIf you would build the voice clips for the numbers into the plugin (only possible if space permits it), you could have different clips for different languages.
13:12:43amiconnOtherwise this will also pose a problem...
13:13:11[IDC]Dragonnot enough space, sorry.
13:13:36[IDC]Dragonbut the number plugin could use the standard, loaded voice.
13:13:52[IDC]Dragonor just return a set of IDs
13:14:14amiconnYes, but for speaking numbers in different languages you net different _sets_ of clips!
13:14:36[IDC]Dragonyou need different voice files anyway?
13:14:50amiconnFor instance, in english only "one" is needed, but for german you need "ein" and "eins"
13:15:16[IDC]Dragonyes, there may be some extra words
13:15:31[IDC]Dragondon't know how bad this is
13:15:56[IDC]Dragongeman anso needs "und" between some numbers
13:15:59[IDC]Dragonalso
13:16:00amiconnSame goes probably for other languages, so this could be quite a number of extra clips. How much space is taken by an empty clip?
13:16:12[IDC]Dragon8 bytes
13:16:21[IDC]Dragonno problem
13:16:41[IDC]Dragonwe have plenty of empty IDs anyway
13:16:56[IDC]Dragonbut you need to build the superset of all
13:17:02_aLFIs it possible to compile code / a kind of plugin in voicefont which will give the id to play for a number ?
13:17:05amiconnSo maybe this is bearable. However, I would still vote for the voice function being incorporated into the voice file.
13:17:18amiconnThis would be much cleaner from a user's point of view.
13:17:50[IDC]Dragonlater...
13:17:55amiconn_aLF: This is one idea we're just discussing.
13:18:00[IDC]Dragonmaybe much later...
13:19:30[IDC]DragonI wonder if there's any existing work on the different language number "algorithms"
13:19:55 Nick [IDC]Dragon is now known as [IDC]Dragon|lunc (~idc-drago@pD9FF8C54.dip.t-dialin.net)
13:20:14[IDC]Dragon|lunclunch makes it too long
13:20:26[IDC]Dragon|luncanyway, afk
13:20:33amiconnI have an algorithm for german, but it needs to be ported to C (from Arexx) - shouldn't be that difficult.
13:30:28 Join AciD [0] (~acid@longchamp44-1-82-67-133-87.fbx.proxad.net)
13:30:29_aLFI found this for french : http://yann.coscoy.free.fr/nombre/info.html (in french...)
13:38:52 Join Galik [0] (~galik@195.137.1.152)
13:47:47***Saving seen data "./dancer.seen"
13:48:06[IDC]Dragon|lunctime for a nap
13:48:10 Quit [IDC]Dragon|lunc ()
14:00
14:13:18 Quit AciD (Read error: 104 (Connection reset by peer))
15:00
15:13:05 Nick c0utta is now known as c0utta{zz} (~c0utta@200.cust47.nsw.dsl.ozemail.com.au)
15:19:54 Part amiconn
15:47:51***Saving seen data "./dancer.seen"
16:00
16:12:51 Join AciD [0] (~acid@longchamp44-1-82-67-133-87.fbx.proxad.net)
17:00
17:00:10 Quit Galik ("Client exiting")
17:05:27 Join pfavr [0] (pfavr@t2o61p235.telia.com)
17:44:05 Quit pfavr ("ChatZilla 0.9.52B [Mozilla rv:1.6/1]")
17:47:52***Saving seen data "./dancer.seen"
17:48:44 Quit AciD (Read error: 104 (Connection reset by peer))
17:52:55 Join Galik [0] (~galik@195.137.1.152)
18:00
18:04:18 Quit Galik ("Client exiting")
18:04:42 Join Galik [0] (~galik@195.137.1.152)
18:04:43 Quit Galik (Read error: 54 (Connection reset by peer))
18:07:12 Join Galik [0] (~galik@195.137.1.152)
18:17:37 Join AciD [0] (~acid@longchamp44-1-82-67-133-87.fbx.proxad.net)
18:41:46 Quit Galik (Remote closed the connection)
18:47:33 Join Galik [0] (~galik@195.137.1.152)
19:00
19:00:20 Nick Ka__ is now known as Ka (~tkirk@pcp261336pcs.howard01.md.comcast.net)
19:00:24 Nick Ka is now known as Ka_ (~tkirk@pcp261336pcs.howard01.md.comcast.net)
19:02:22 Quit Galik ("Client exiting")
19:47:57***Saving seen data "./dancer.seen"
20:00
20:05:31 Quit jkerman (Read error: 104 (Connection reset by peer))
20:06:07 Join jkerman [0] (~jkerman@jkhouse2.jvlnet.com)
20:56:19 Quit methangas (" HydraIRC -> http://www.hydrairc.com <- IRC has never been so cool")
20:58:41 Join methangas [0] (methangas@0x50a46197.virnxx10.adsl-dhcp.tele.dk)
21:00
21:06:29 Quit joshn_ (Operation timed out)
21:06:52 Quit joshn__ (Operation timed out)
21:48:01***Saving seen data "./dancer.seen"
22:00
22:08:12 Join diddystar5 [0] (lee@IC87.library.oregonstate.edu)
22:17:00 Quit AciD (Read error: 104 (Connection reset by peer))
22:17:42 Join _aLEX [0] (Alexandre@mutualite-3-82-67-66-128.fbx.proxad.net)
22:18:22 Quit _aLF (Read error: 54 (Connection reset by peer))
22:25:47 Quit _aLEX ("Leaving")
23:00
23:03:50diddystar5someone needs to fix the sim
23:03:54diddystar5it crashes
23:14:33 Join amiconn [0] (~jens@pD95D134D.dip.t-dialin.net)
23:18:26 Quit joshn (Connection reset by peer)
23:19:25 Join track [0] (74d57721@ACBA5A05.ipt.aol.com)
23:19:29 Quit track (Client Quit)
23:48:04***Saving seen data "./dancer.seen"

Previous day | Next day